diff --git a/PPOCRLabel/PPOCRLabel.py b/PPOCRLabel/PPOCRLabel.py index 7c7802a73ed32680142f8119b10a0393d1fab9cc..1b902484e2352fda27efa787be39ddb52150f1c9 100644 --- a/PPOCRLabel/PPOCRLabel.py +++ b/PPOCRLabel/PPOCRLabel.py @@ -21,12 +21,13 @@ import os.path import platform import subprocess import sys +import xlrd from functools import partial from PyQt5.QtCore import QSize, Qt, QPoint, QByteArray, QTimer, QFileInfo, QPointF, QProcess from PyQt5.QtGui import QImage, QCursor, QPixmap, QImageReader from PyQt5.QtWidgets import QMainWindow, QListWidget, QVBoxLayout, QToolButton, QHBoxLayout, QDockWidget, QWidget, \ - QSlider, QGraphicsOpacityEffect, QMessageBox, QListView, QScrollArea, QWidgetAction, QApplication, QLabel, \ + QSlider, QGraphicsOpacityEffect, QMessageBox, QListView, QScrollArea, QWidgetAction, QApplication, QLabel, QGridLayout, \ QFileDialog, QListWidgetItem, QComboBox, QDialog __dir__ = os.path.dirname(os.path.abspath(__file__)) @@ -36,7 +37,7 @@ sys.path.append(os.path.abspath(os.path.join(__dir__, '../..'))) sys.path.append(os.path.abspath(os.path.join(__dir__, '../PaddleOCR'))) sys.path.append("..") -from paddleocr import PaddleOCR +from paddleocr import PaddleOCR, PPStructure from libs.constants import * from libs.utils import * from libs.labelColor import label_colormap @@ -100,9 +101,15 @@ class MainWindow(QMainWindow): use_gpu=gpu, lang=lang, show_log=False) + self.table_ocr = PPStructure(use_pdserving=False, + use_gpu=gpu, + lang=lang, + layout=False, + show_log=False) if os.path.exists('./data/paddle.png'): result = self.ocr.ocr('./data/paddle.png', cls=True, det=True) + result = self.table_ocr('./data/paddle.png', return_ocr_result_in_table=True) # For loading all image under a directory self.mImgList = [] @@ -196,16 +203,25 @@ class MainWindow(QMainWindow): self.reRecogButton.setIcon(newIcon('reRec', 30)) self.reRecogButton.setToolButtonStyle(Qt.ToolButtonTextBesideIcon) + self.tableRecButton = QToolButton() + self.tableRecButton.setToolButtonStyle(Qt.ToolButtonTextBesideIcon) + self.newButton = QToolButton() self.newButton.setToolButtonStyle(Qt.ToolButtonTextBesideIcon) + self.createpolyButton = QToolButton() + self.createpolyButton.setToolButtonStyle(Qt.ToolButtonTextBesideIcon) + self.SaveButton = QToolButton() self.SaveButton.setToolButtonStyle(Qt.ToolButtonTextBesideIcon) self.DelButton = QToolButton() self.DelButton.setToolButtonStyle(Qt.ToolButtonTextBesideIcon) - leftTopToolBox = QHBoxLayout() - leftTopToolBox.addWidget(self.newButton) - leftTopToolBox.addWidget(self.reRecogButton) + leftTopToolBox = QGridLayout() + leftTopToolBox.addWidget(self.newButton, 0, 0, 1, 1) + leftTopToolBox.addWidget(self.createpolyButton, 0, 1, 1, 1) + leftTopToolBox.addWidget(self.reRecogButton, 1, 0, 1, 1) + leftTopToolBox.addWidget(self.tableRecButton, 1, 1, 1, 1) + leftTopToolBoxContainer = QWidget() leftTopToolBoxContainer.setLayout(leftTopToolBox) listLayout.addWidget(leftTopToolBoxContainer) @@ -446,13 +462,22 @@ class MainWindow(QMainWindow): 'Ctrl+R', 'reRec', getStr('singleRe'), enabled=False) createpoly = action(getStr('creatPolygon'), self.createPolygon, - 'q', 'new', getStr('creatPolygon'), enabled=True) + 'q', 'new', getStr('creatPolygon'), enabled=False) + + tableRec = action(getStr('TableRecognition'), self.TableRecognition, + '', 'Auto', getStr('TableRecognition'), enabled=False) + + cellreRec = action(getStr('cellreRecognition'), self.cellreRecognition, + '', 'reRec', getStr('cellreRecognition'), enabled=False) saveRec = action(getStr('saveRec'), self.saveRecResult, '', 'save', getStr('saveRec'), enabled=False) saveLabel = action(getStr('saveLabel'), self.saveLabelFile, # 'Ctrl+S', 'save', getStr('saveLabel'), enabled=False) + + exportJSON = action(getStr('exportJSON'), self.exportJSON, + '', 'save', getStr('exportJSON'), enabled=False) undoLastPoint = action(getStr("undoLastPoint"), self.canvas.undoLastPoint, 'Ctrl+Z', "undo", getStr("undoLastPoint"), enabled=False) @@ -474,10 +499,12 @@ class MainWindow(QMainWindow): self.editButton.setDefaultAction(edit) self.newButton.setDefaultAction(create) + self.createpolyButton.setDefaultAction(createpoly) self.DelButton.setDefaultAction(deleteImg) self.SaveButton.setDefaultAction(save) self.AutoRecognition.setDefaultAction(AutoRec) self.reRecogButton.setDefaultAction(reRec) + self.tableRecButton.setDefaultAction(tableRec) # self.preButton.setDefaultAction(openPrevImg) # self.nextButton.setDefaultAction(openNextImg) @@ -523,25 +550,25 @@ class MainWindow(QMainWindow): # Store actions for further handling. self.actions = struct(save=save, resetAll=resetAll, deleteImg=deleteImg, - lineColor=color1, create=create, delete=delete, edit=edit, copy=copy, - saveRec=saveRec, singleRere=singleRere, AutoRec=AutoRec, reRec=reRec, + lineColor=color1, create=create, createpoly=createpoly, tableRec=tableRec, delete=delete, edit=edit, copy=copy, + saveRec=saveRec, singleRere=singleRere, AutoRec=AutoRec, reRec=reRec, cellreRec=cellreRec, createMode=createMode, editMode=editMode, shapeLineColor=shapeLineColor, shapeFillColor=shapeFillColor, zoom=zoom, zoomIn=zoomIn, zoomOut=zoomOut, zoomOrg=zoomOrg, fitWindow=fitWindow, fitWidth=fitWidth, zoomActions=zoomActions, saveLabel=saveLabel, change_cls=change_cls, undo=undo, undoLastPoint=undoLastPoint, open_dataset_dir=open_dataset_dir, - rotateLeft=rotateLeft, rotateRight=rotateRight, lock=lock, - fileMenuActions=(opendir, open_dataset_dir, saveLabel, resetAll, quit), + rotateLeft=rotateLeft, rotateRight=rotateRight, lock=lock, exportJSON=exportJSON, + fileMenuActions=(opendir, open_dataset_dir, saveLabel, exportJSON, resetAll, quit), beginner=(), advanced=(), - editMenu=(createpoly, edit, copy, delete, singleRere, None, undo, undoLastPoint, + editMenu=(createpoly, edit, copy, delete, singleRere, cellreRec, None, undo, undoLastPoint, None, rotateLeft, rotateRight, None, color1, self.drawSquaresOption, lock, None, change_cls), beginnerContext=( - create, edit, copy, delete, singleRere, rotateLeft, rotateRight, lock, change_cls), + create, createpoly, edit, copy, delete, singleRere, cellreRec, rotateLeft, rotateRight, lock, change_cls), advancedContext=(createMode, editMode, edit, copy, delete, shapeLineColor, shapeFillColor), - onLoadActive=(create, createMode, editMode), + onLoadActive=(create, createpoly, createMode, editMode), onShapesPresent=(hideAll, showAll)) # menus @@ -574,7 +601,7 @@ class MainWindow(QMainWindow): self.autoSaveOption.triggered.connect(self.autoSaveFunc) addActions(self.menus.file, - (opendir, open_dataset_dir, None, saveLabel, saveRec, self.autoSaveOption, None, resetAll, deleteImg, + (opendir, open_dataset_dir, None, saveLabel, saveRec, exportJSON, self.autoSaveOption, None, resetAll, deleteImg, quit)) addActions(self.menus.help, (showKeys, showSteps, showInfo)) @@ -585,7 +612,7 @@ class MainWindow(QMainWindow): zoomIn, zoomOut, zoomOrg, None, fitWindow, fitWidth)) - addActions(self.menus.autolabel, (AutoRec, reRec, alcm, None, help)) + addActions(self.menus.autolabel, (AutoRec, reRec, cellreRec, alcm, None, help)) self.menus.file.aboutToShow.connect(self.updateFileMenu) @@ -695,6 +722,7 @@ class MainWindow(QMainWindow): self.dirty = False self.actions.save.setEnabled(False) self.actions.create.setEnabled(True) + self.actions.createpoly.setEnabled(True) def toggleActions(self, value=True): """Enable/Disable widgets which depend on an opened image.""" @@ -780,6 +808,7 @@ class MainWindow(QMainWindow): assert self.beginner() self.canvas.setEditing(False) self.actions.create.setEnabled(False) + self.actions.createpoly.setEnabled(False) self.canvas.fourpoint = False def createPolygon(self): @@ -787,10 +816,10 @@ class MainWindow(QMainWindow): self.canvas.setEditing(False) self.canvas.fourpoint = True self.actions.create.setEnabled(False) + self.actions.createpoly.setEnabled(False) self.actions.undoLastPoint.setEnabled(True) def rotateImg(self, filename, k, _value): - self.actions.rotateRight.setEnabled(_value) pix = cv2.imread(filename) pix = np.rot90(pix, k) @@ -831,6 +860,7 @@ class MainWindow(QMainWindow): self.canvas.setEditing(True) self.canvas.restoreCursor() self.actions.create.setEnabled(True) + self.actions.createpoly.setEnabled(True) def toggleDrawMode(self, edit=True): self.canvas.setEditing(edit) @@ -987,11 +1017,21 @@ class MainWindow(QMainWindow): if len(self.canvas.selectedShapes) == 1 and self.keyList.count() > 0: selected_key_item_row = self.keyList.findItemsByLabel(self.canvas.selectedShapes[0].key_cls, get_row=True) + if isinstance(selected_key_item_row, list) and len(selected_key_item_row) == 0: + key_text = self.canvas.selectedShapes[0].key_cls + item = self.keyList.createItemFromLabel(key_text) + self.keyList.addItem(item) + rgb = self._get_rgb_by_label(key_text, self.kie_mode) + self.keyList.setItemLabel(item, key_text, rgb) + selected_key_item_row = self.keyList.findItemsByLabel(self.canvas.selectedShapes[0].key_cls, + get_row=True) + self.keyList.setCurrentRow(selected_key_item_row) self._noSelectionSlot = False n_selected = len(selected_shapes) self.actions.singleRere.setEnabled(n_selected) + self.actions.cellreRec.setEnabled(n_selected) self.actions.delete.setEnabled(n_selected) self.actions.copy.setEnabled(n_selected) self.actions.edit.setEnabled(n_selected == 1) @@ -1216,6 +1256,7 @@ class MainWindow(QMainWindow): if self.beginner(): # Switch to edit mode. self.canvas.setEditing(True) self.actions.create.setEnabled(True) + self.actions.createpoly.setEnabled(True) self.actions.undoLastPoint.setEnabled(False) self.actions.undo.setEnabled(True) else: @@ -1654,8 +1695,10 @@ class MainWindow(QMainWindow): self.haveAutoReced = False self.AutoRecognition.setEnabled(True) self.reRecogButton.setEnabled(True) + self.tableRecButton.setEnabled(True) self.actions.AutoRec.setEnabled(True) self.actions.reRec.setEnabled(True) + self.actions.tableRec.setEnabled(True) self.actions.open_dataset_dir.setEnabled(True) self.actions.rotateLeft.setEnabled(True) self.actions.rotateRight.setEnabled(True) @@ -1755,6 +1798,7 @@ class MainWindow(QMainWindow): self.openNextImg() self.actions.saveRec.setEnabled(True) self.actions.saveLabel.setEnabled(True) + self.actions.exportJSON.setEnabled(True) elif mode == 'Auto': if annotationFilePath and self.saveLabels(annotationFilePath, mode=mode): @@ -2081,6 +2125,280 @@ class MainWindow(QMainWindow): self.singleLabel(shape) self.setDirty() + def TableRecognition(self): + ''' + Table Recegnition + ''' + from paddleocr.ppstructure.table.predict_table import to_excel + + import time + + start = time.time() + img = cv2.imread(self.filePath) + res = self.table_ocr(img, return_ocr_result_in_table=True) + + TableRec_excel_dir = self.lastOpenDir + '/tableRec_excel_output/' + os.makedirs(TableRec_excel_dir, exist_ok=True) + filename, _ = os.path.splitext(os.path.basename(self.filePath)) + + excel_path = TableRec_excel_dir + '{}.xlsx'.format(filename) + + if res is None: + msg = 'Can not recognise the table in ' + self.filePath + '. Please change manually' + QMessageBox.information(self, "Information", msg) + to_excel('', excel_path) # create an empty excel + return + + # save res + # ONLY SUPPORT ONE TABLE in one image + hasTable = False + for region in res: + if region['type'] == 'Table': + if region['res']['boxes'] is None: + msg = 'Can not recognise the detection box in ' + self.filePath + '. Please change manually' + QMessageBox.information(self, "Information", msg) + to_excel('', excel_path) # create an empty excel + return + hasTable = True + # save table ocr result on PPOCRLabel + # clear all old annotaions before saving result + self.itemsToShapes.clear() + self.shapesToItems.clear() + self.itemsToShapesbox.clear() # ADD + self.shapesToItemsbox.clear() + self.labelList.clear() + self.BoxList.clear() + self.result_dic = [] + self.result_dic_locked = [] + + shapes = [] + result_len = len(region['res']['boxes']) + for i in range(result_len): + bbox = np.array(region['res']['boxes'][i]) + rec_text = region['res']['rec_res'][i][0] + + # polys to rectangles + x1, y1 = np.min(bbox[:, 0]), np.min(bbox[:, 1]) + x2, y2 = np.max(bbox[:, 0]), np.max(bbox[:, 1]) + rext_bbox = [[x1, y1], [x2, y1], [x2, y2], [x1, y2]] + + # save bbox to shape + shape = Shape(label=rec_text, line_color=DEFAULT_LINE_COLOR, key_cls=None) + for point in rext_bbox: + x, y = point + # Ensure the labels are within the bounds of the image. + # If not, fix them. + x, y, snapped = self.canvas.snapPointToCanvas(x, y) + shape.addPoint(QPointF(x, y)) + shape.difficult = False + # shape.locked = False + shape.close() + self.addLabel(shape) + shapes.append(shape) + self.setDirty() + self.canvas.loadShapes(shapes) + + # save HTML result to excel + try: + to_excel(region['res']['html'], excel_path) + except: + print('Can not save excel file, maybe Permission denied (.xlsx is being occupied)') + break + + if not hasTable: + msg = 'Can not recognise the table in ' + self.filePath + '. Please change manually' + QMessageBox.information(self, "Information", msg) + to_excel('', excel_path) # create an empty excel + return + + # automatically open excel annotation file + if platform.system() == 'Windows': + try: + import win32com.client + except: + print("CANNOT OPEN .xlsx. It could be one of the following reasons: " \ + "Only support Windows | No python win32com") + + try: + xl = win32com.client.Dispatch("Excel.Application") + xl.Visible = True + xl.Workbooks.Open(excel_path) + # excelEx = "You need to show the excel executable at this point" + # subprocess.Popen([excelEx, excel_path]) + + # os.startfile(excel_path) + except: + print("CANNOT OPEN .xlsx. It could be the following reasons: " \ + ".xlsx is not existed") + else: + os.system('open ' + os.path.normpath(excel_path)) + + print('time cost: ', time.time() - start) + + def cellreRecognition(self): + ''' + re-recognise text in a cell + ''' + img = cv2.imread(self.filePath) + for shape in self.canvas.selectedShapes: + box = [[int(p.x()), int(p.y())] for p in shape.points] + + if len(box) > 4: + box = self.gen_quad_from_poly(np.array(box)) + assert len(box) == 4 + + # pad around bbox for better text recognition accuracy + _box = boxPad(box, img.shape, 6) + img_crop = get_rotate_crop_image(img, np.array(_box, np.float32)) + if img_crop is None: + msg = 'Can not recognise the detection box in ' + self.filePath + '. Please change manually' + QMessageBox.information(self, "Information", msg) + return + + # merge the text result in the cell + texts = '' + probs = 0. # the probability of the cell is avgerage prob of every text box in the cell + bboxes = self.ocr.ocr(img_crop, det=True, rec=False, cls=False) + if len(bboxes) > 0: + bboxes.reverse() # top row text at first + for _bbox in bboxes: + patch = get_rotate_crop_image(img_crop, np.array(_bbox, np.float32)) + rec_res = self.ocr.ocr(patch, det=False, rec=True, cls=False) + text = rec_res[0][0] + if text != '': + texts += text + (' ' if text[0].isalpha() else '') # add space between english word + probs += rec_res[0][1] + probs = probs / len(bboxes) + result = [(texts.strip(), probs)] + + if result[0][0] != '': + result.insert(0, box) + print('result in reRec is ', result) + if result[1][0] == shape.label: + print('label no change') + else: + shape.label = result[1][0] + else: + print('Can not recognise the box') + if self.noLabelText == shape.label: + print('label no change') + else: + shape.label = self.noLabelText + self.singleLabel(shape) + self.setDirty() + + def exportJSON(self): + ''' + export PPLabel and CSV to JSON (PubTabNet) + ''' + import pandas as pd + from libs.dataPartitionDialog import DataPartitionDialog + + # data partition user input + partitionDialog = DataPartitionDialog(parent=self) + partitionDialog.exec() + if partitionDialog.getStatus() == False: + return + + # automatically save annotations + self.saveFilestate() + self.savePPlabel(mode='auto') + + # load box annotations + labeldict = {} + if not os.path.exists(self.PPlabelpath): + msg = 'ERROR, Can not find Label.txt' + QMessageBox.information(self, "Information", msg) + return + else: + with open(self.PPlabelpath, 'r', encoding='utf-8') as f: + data = f.readlines() + for each in data: + file, label = each.split('\t') + if label: + label = label.replace('false', 'False') + label = label.replace('true', 'True') + labeldict[file] = eval(label) + else: + labeldict[file] = [] + + # if len(labeldict) != len(csv_paths): + # msg = 'ERROR, box label and excel label are not in the same number\n' + \ + # 'box label: ' + str(len(labeldict)) + '\n' + \ + # 'excel label: ' + str(len(csv_paths)) + '\n' + \ + # 'Please check the label.txt and tableRec_excel_output\n' + # QMessageBox.information(self, "Information", msg) + # return + train_split, val_split, test_split = partitionDialog.getDataPartition() + # check validate + if train_split + val_split + test_split > 100: + msg = "The sum of training, validation and testing data should be less than 100%" + QMessageBox.information(self, "Information", msg) + return + print(train_split, val_split, test_split) + train_split, val_split, test_split = float(train_split) / 100., float(val_split) / 100., float(test_split) / 100. + train_id = int(len(labeldict) * train_split) + val_id = int(len(labeldict) * (train_split + val_split)) + print('Data partition: train:', train_id, + 'validation:', val_id - train_id, + 'test:', len(labeldict) - val_id) + + TableRec_excel_dir = os.path.join(self.lastOpenDir, 'tableRec_excel_output') + json_results = [] + imgid = 0 + for image_path in labeldict.keys(): + # load csv annotations + filename, _ = os.path.splitext(os.path.basename(image_path)) + csv_path = os.path.join(TableRec_excel_dir, filename + '.xlsx') + if not os.path.exists(csv_path): + msg = 'ERROR, Can not find ' + csv_path + QMessageBox.information(self, "Information", msg) + return + + # read xlsx file, convert to HTML + # xd = pd.ExcelFile(csv_path) + # df = xd.parse() + # structure = df.to_html(index = False) + excel = xlrd.open_workbook(csv_path) + sheet0 = excel.sheet_by_index(0) # only sheet 0 + merged_cells = sheet0.merged_cells # (0,1,1,3) start row, end row, start col, end col + + html_list = [['td'] * sheet0.ncols for i in range(sheet0.nrows)] + + for merged in merged_cells: + html_list = expand_list(merged, html_list) + + token_list = convert_token(html_list) + + + # load box annotations + cells = [] + for anno in labeldict[image_path]: + tokens = list(anno['transcription']) + obb = anno['points'] + hbb = OBB2HBB(np.array(obb)).tolist() + cells.append({'tokens': tokens, 'bbox': hbb}) + + # data split + if imgid < train_id: + split = 'train' + elif imgid < val_id: + split = 'val' + else: + split = 'test' + + # save dict + html = {'structure': {'tokens': token_list}, 'cell': cells} + json_results.append({'filename': os.path.basename(image_path), 'split': split, 'imgid': imgid, 'html': html}) + imgid += 1 + + # save json + with open("{}/annotation.json".format(self.lastOpenDir), "w", encoding='utf-8') as fid: + fid.write(json.dumps(json_results, ensure_ascii=False)) + + msg = 'JSON sucessfully saved in {}/annotation.json'.format(self.lastOpenDir) + QMessageBox.information(self, "Information", msg) + def autolcm(self): vbox = QVBoxLayout() hbox = QHBoxLayout() @@ -2120,6 +2438,12 @@ class MainWindow(QMainWindow): del self.ocr self.ocr = PaddleOCR(use_pdserving=False, use_angle_cls=True, det=True, cls=True, use_gpu=False, lang=lg_idx[self.comboBox.currentText()]) + del self.table_ocr + self.table_ocr = PPStructure(use_pdserving=False, + use_gpu=False, + lang=lg_idx[self.comboBox.currentText()], + layout=False, + show_log=False) self.dialog.close() def cancel(self): @@ -2138,6 +2462,7 @@ class MainWindow(QMainWindow): self.fileStatedict[file] = 1 self.actions.saveLabel.setEnabled(True) self.actions.saveRec.setEnabled(True) + self.actions.exportJSON.setEnabled(True) def saveFilestate(self): with open(self.fileStatepath, 'w', encoding='utf-8') as f: diff --git a/PPOCRLabel/libs/dataPartitionDialog.py b/PPOCRLabel/libs/dataPartitionDialog.py new file mode 100644 index 0000000000000000000000000000000000000000..33bd491552fe773bd07020d82f7ea9bab76e7557 --- /dev/null +++ b/PPOCRLabel/libs/dataPartitionDialog.py @@ -0,0 +1,113 @@ +try: + from PyQt5.QtGui import * + from PyQt5.QtCore import * + from PyQt5.QtWidgets import * +except ImportError: + from PyQt4.QtGui import * + from PyQt4.QtCore import * + +from libs.utils import newIcon + +import time +import datetime +import json +import cv2 +import numpy as np + + +BB = QDialogButtonBox + +class DataPartitionDialog(QDialog): + def __init__(self, parent=None): + super().__init__() + self.parnet = parent + self.title = 'DATA PARTITION' + + self.train_ratio = 70 + self.val_ratio = 15 + self.test_ratio = 15 + + self.initUI() + + def initUI(self): + self.setWindowTitle(self.title) + self.setWindowModality(Qt.ApplicationModal) + + self.flag_accept = True + + if self.parnet.lang == 'ch': + msg = "导出JSON前请保存所有图像的标注且关闭EXCEL!" + else: + msg = "Please save all the annotations and close the EXCEL before exporting JSON!" + + info_msg = QLabel(msg, self) + info_msg.setWordWrap(True) + info_msg.setStyleSheet("color: red") + info_msg.setFont(QFont('Arial', 12)) + + train_lbl = QLabel('Train split: ', self) + train_lbl.setFont(QFont('Arial', 15)) + val_lbl = QLabel('Valid split: ', self) + val_lbl.setFont(QFont('Arial', 15)) + test_lbl = QLabel('Test split: ', self) + test_lbl.setFont(QFont('Arial', 15)) + + self.train_input = QLineEdit(self) + self.train_input.setFont(QFont('Arial', 15)) + self.val_input = QLineEdit(self) + self.val_input.setFont(QFont('Arial', 15)) + self.test_input = QLineEdit(self) + self.test_input.setFont(QFont('Arial', 15)) + + self.train_input.setText(str(self.train_ratio)) + self.val_input.setText(str(self.val_ratio)) + self.test_input.setText(str(self.test_ratio)) + + validator = QIntValidator(0, 100) + self.train_input.setValidator(validator) + self.val_input.setValidator(validator) + self.test_input.setValidator(validator) + + gridlayout = QGridLayout() + gridlayout.addWidget(info_msg, 0, 0, 1, 2) + gridlayout.addWidget(train_lbl, 1, 0) + gridlayout.addWidget(val_lbl, 2, 0) + gridlayout.addWidget(test_lbl, 3, 0) + gridlayout.addWidget(self.train_input, 1, 1) + gridlayout.addWidget(self.val_input, 2, 1) + gridlayout.addWidget(self.test_input, 3, 1) + + bb = BB(BB.Ok | BB.Cancel, Qt.Horizontal, self) + bb.button(BB.Ok).setIcon(newIcon('done')) + bb.button(BB.Cancel).setIcon(newIcon('undo')) + bb.accepted.connect(self.validate) + bb.rejected.connect(self.cancel) + gridlayout.addWidget(bb, 4, 0, 1, 2) + + self.setLayout(gridlayout) + + self.show() + + def validate(self): + self.flag_accept = True + self.accept() + + def cancel(self): + self.flag_accept = False + self.reject() + + def getStatus(self): + return self.flag_accept + + def getDataPartition(self): + self.train_ratio = int(self.train_input.text()) + self.val_ratio = int(self.val_input.text()) + self.test_ratio = int(self.test_input.text()) + + return self.train_ratio, self.val_ratio, self.test_ratio + + def closeEvent(self, event): + self.flag_accept = False + self.reject() + + diff --git a/PPOCRLabel/libs/utils.py b/PPOCRLabel/libs/utils.py index 2510520caa8048d7787d7c8f65df2885d76026f7..bf54700488e285da8a89b2648a17e0e1a7341b60 100644 --- a/PPOCRLabel/libs/utils.py +++ b/PPOCRLabel/libs/utils.py @@ -161,6 +161,77 @@ def get_rotate_crop_image(img, points): print(e) +def boxPad(box, imgShape, pad : int) -> np.array: + """ + Pad a box with [pad] pixels on each side. + """ + box = np.array(box, dtype=np.int32) + box[0][0], box[0][1] = box[0][0] - pad, box[0][1] - pad + box[1][0], box[1][1] = box[1][0] + pad, box[1][1] - pad + box[2][0], box[2][1] = box[2][0] + pad, box[2][1] + pad + box[3][0], box[3][1] = box[3][0] - pad, box[3][1] + pad + h, w, _ = imgShape + box[:,0] = np.clip(box[:,0], 0, w) + box[:,1] = np.clip(box[:,1], 0, h) + return box + + +def OBB2HBB(obb) -> np.array: + """ + Convert Oriented Bounding Box to Horizontal Bounding Box. + """ + hbb = np.zeros(4, dtype=np.int32) + hbb[0] = min(obb[:, 0]) + hbb[1] = min(obb[:, 1]) + hbb[2] = max(obb[:, 0]) + hbb[3] = max(obb[:, 1]) + return hbb + + +def expand_list(merged, html_list): + ''' + Fill blanks according to merged cells + ''' + sr, er, sc, ec = merged + for i in range(sr, er): + for j in range(sc, ec): + html_list[i][j] = None + html_list[sr][sc] = '' + if ec - sc > 1: + html_list[sr][sc] += " colspan={}".format(ec - sc) + if er - sr > 1: + html_list[sr][sc] += " rowspan={}".format(er - sr) + return html_list + + +def convert_token(html_list): + ''' + Convert raw html to label format + ''' + token_list = [""] + # final html list: + for row in html_list: + token_list.append("") + for col in row: + if col == None: + continue + elif col == 'td': + token_list.extend(["", ""]) + else: + token_list.append("", ""]) + token_list.append("") + token_list.append("") + + return token_list + + def stepsInfo(lang='en'): if lang == 'ch': msg = "1. 安装与运行：使用上述命令安装与运行程序。\n" \ diff --git a/PPOCRLabel/resources/strings/strings-en.properties b/PPOCRLabel/resources/strings/strings-en.properties index 3c4eda65a32e1048405041667ba61bdb639bfd7b..7ba9af4c33ebe8130b5ce529e01270361c5f11d6 100644 --- a/PPOCRLabel/resources/strings/strings-en.properties +++ b/PPOCRLabel/resources/strings/strings-en.properties @@ -84,7 +84,7 @@ mhelp=Help iconList=Icon List detectionBoxposition=Detection box position recognitionResult=Recognition result -creatPolygon=Create Quadrilateral +creatPolygon=Create PolygonBox rotateLeft=Left turn 90 degrees rotateRight=Right turn 90 degrees drawSquares=Draw Squares @@ -110,3 +110,6 @@ lockBoxDetail=Lock selected box/Unlock all box keyListTitle=Key List keyDialogTip=Enter object label keyChange=Change Box Key +TableRecognition=Table Recognition +cellreRecognition=Cell Re-Recognition +exportJSON=export JSON(PubTabNet) diff --git a/PPOCRLabel/resources/strings/strings-zh-CN.properties b/PPOCRLabel/resources/strings/strings-zh-CN.properties index a7c30368b87354cbae81b2cdead8ad31b2a8c1eb..308974ef29fac367bccd5f0e97926b2087a9f8df 100644 --- a/PPOCRLabel/resources/strings/strings-zh-CN.properties +++ b/PPOCRLabel/resources/strings/strings-zh-CN.properties @@ -84,7 +84,7 @@ mhelp=帮助 iconList=缩略图 detectionBoxposition=检测框位置 recognitionResult=识别结果 -creatPolygon=四点标注 +creatPolygon=多边形标注 drawSquares=正方形标注 rotateLeft=图片左旋转90度 rotateRight=图片右旋转90度 @@ -109,4 +109,7 @@ lockBox=锁定框/解除锁定框 lockBoxDetail=若当前没有框处于锁定状态则锁定选中的框，若存在锁定框则解除所有锁定框的锁定状态 keyListTitle=关键词列表 keyDialogTip=请输入类型名称 -keyChange=更改Box关键字类别 \ No newline at end of file +keyChange=更改Box关键字类别 +TableRecognition=表格识别 +cellreRecognition=单元格重识别 +exportJSON=导出表格JSON标注 \ No newline at end of file diff --git a/README.md b/README.md index 511b0f299cdfeeef9e623240e9ab9b2c5364e006..835b1e2509ebca6f6d0dd71a53a7ec02a147efcf 100644 --- a/README.md +++ b/README.md @@ -30,8 +30,6 @@ PaddleOCR aims to create multilingual, awesome, leading, and practical OCR tools PaddleOCR support a variety of cutting-edge algorithms related to OCR, and developed industrial featured models/solution [PP-OCR](./doc/doc_en/ppocr_introduction_en.md) and [PP-Structure](./ppstructure/README.md) on this basis, and get through the whole process of data production, model training, compression, inference and deployment. -PaddleOCR also supports metric and model logging during training to [VisualDL](https://www.paddlepaddle.org.cn/documentation/docs/en/guides/03_VisualDL/visualdl_usage_en.html) and [Weights & Biases](https://docs.wandb.ai/). - ![](./doc/features_en.png) > It is recommended to start with the “quick experience” in the document tutorial diff --git a/deploy/pdserving/config.yml b/deploy/pdserving/config.yml index 19cd9325ee8b241fd591678b9ba6452de9bec025..8014cbd362461ead5d065f96a50eb3031a60fa67 100644 --- a/deploy/pdserving/config.yml +++ b/deploy/pdserving/config.yml @@ -36,8 +36,8 @@ op: #det模型路径 model_config: ./ppocr_det_v3_serving - #Fetch结果列表，以client_config中fetch_var的alias_name为准 - fetch_list: ["sigmoid_0.tmp_0"] + #Fetch结果列表，以client_config中fetch_var的alias_name为准，不设置默认取全部输出变量 + #fetch_list: ["sigmoid_0.tmp_0"] #计算硬件ID，当devices为""或不写时为CPU预测；当devices为"0", "0,1,2"时为GPU预测，表示使用的GPU卡 devices: "0" @@ -62,8 +62,8 @@ op: #rec模型路径 model_config: ./ppocr_rec_v3_serving - #Fetch结果列表，以client_config中fetch_var的alias_name为准 - fetch_list: ["softmax_5.tmp_0"] + #Fetch结果列表，以client_config中fetch_var的alias_name为准, 不设置默认取全部输出变量 + #fetch_list: #计算硬件ID，当devices为""或不写时为CPU预测；当devices为"0", "0,1,2"时为GPU预测，表示使用的GPU卡 devices: "0" diff --git a/deploy/pdserving/ocr_reader.py b/deploy/pdserving/ocr_reader.py index 6a2d57b679d69ab11ac6f0fd74c47a342b391545..75f0f3d5c3aea488f82ec01a72e20310663d565b 100644 --- a/deploy/pdserving/ocr_reader.py +++ b/deploy/pdserving/ocr_reader.py @@ -393,7 +393,7 @@ class OCRReader(object): return norm_img_batch[0] def postprocess(self, outputs, with_score=False): - preds = outputs["softmax_5.tmp_0"] + preds = list(outputs.values())[0] try: preds = preds.numpy() except: @@ -404,8 +404,11 @@ class OCRReader(object): preds_idx, preds_prob, is_remove_duplicate=True) return text -from argparse import ArgumentParser,RawDescriptionHelpFormatter + +from argparse import ArgumentParser, RawDescriptionHelpFormatter import yaml + + class ArgsParser(ArgumentParser): def __init__(self): super(ArgsParser, self).__init__( @@ -441,16 +444,16 @@ class ArgsParser(ArgumentParser): s = s.strip() k, v = s.split('=') v = self._parse_helper(v) - print(k,v, type(v)) + print(k, v, type(v)) cur = config parent = cur for kk in k.split("."): if kk not in cur: - cur[kk] = {} - parent = cur - cur = cur[kk] + cur[kk] = {} + parent = cur + cur = cur[kk] else: - parent = cur - cur = cur[kk] + parent = cur + cur = cur[kk] parent[k.split(".")[-1]] = v - return config \ No newline at end of file + return config diff --git a/deploy/pdserving/web_service.py b/deploy/pdserving/web_service.py index 98e2dfba2f5abd3fc36bf3743b23f7eb7be3b9c4..f05806ce030238144568a3ca137798a9132027e4 100644 --- a/deploy/pdserving/web_service.py +++ b/deploy/pdserving/web_service.py @@ -56,7 +56,7 @@ class DetOp(Op): return {"x": det_img[np.newaxis, :].copy()}, False, None, "" def postprocess(self, input_dicts, fetch_dict, data_id, log_id): - det_out = fetch_dict["sigmoid_0.tmp_0"] + det_out = list(fetch_dict.values())[0] ratio_list = [ float(self.new_h) / self.ori_h, float(self.new_w) / self.ori_w ] diff --git a/deploy/pdserving/web_service_det.py b/deploy/pdserving/web_service_det.py index 7584608a9fed4bea93caa5c814c0450566696d56..4a62ab861d8338194da826cdcea2d42de189c994 100644 --- a/deploy/pdserving/web_service_det.py +++ b/deploy/pdserving/web_service_det.py @@ -55,7 +55,7 @@ class DetOp(Op): return {"x": det_img[np.newaxis, :].copy()}, False, None, "" def postprocess(self, input_dicts, fetch_dict, data_id, log_id): - det_out = fetch_dict["sigmoid_0.tmp_0"] + det_out = list(fetch_dict.values())[0] ratio_list = [ float(self.new_h) / self.ori_h, float(self.new_w) / self.ori_w ] diff --git a/doc/doc_ch/PP-OCRv3_introduction.md b/doc/doc_ch/PP-OCRv3_introduction.md index c7ef4f01569437959e310f2301d250c1660113a7..78e517d1e66696ee350123e8002252c4b2d7db0e 100644 --- a/doc/doc_ch/PP-OCRv3_introduction.md +++ b/doc/doc_ch/PP-OCRv3_introduction.md @@ -110,14 +110,15 @@ PP-OCRv3识别模型从网络结构、训练策略、数据增广等多个方面 |-----|-----|--------|----| --- | | 01 | PP-OCRv2 | 8M | 74.8% | 8.54ms | | 02 | SVTR_Tiny | 21M | 80.1% | 97ms | -| 03 | SVTR_LCNet | 12M | 71.9% | 6.6ms | -| 04 | + GTC | 12M | 75.8% | 7.6ms | -| 05 | + TextConAug | 12M | 76.3% | 7.6ms | -| 06 | + TextRotNet | 12M | 76.9% | 7.6ms | -| 07 | + UDML | 12M | 78.4% | 7.6ms | -| 08 | + UIM | 12M | 79.4% | 7.6ms | +| 03 | SVTR_LCNet(h32) | 12M | 71.9% | 6.6ms | +| 04 | SVTR_LCNet(h48) | 12M | 73.98% | 7.6ms | +| 05 | + GTC | 12M | 75.8% | 7.6ms | +| 06 | + TextConAug | 12M | 76.3% | 7.6ms | +| 07 | + TextRotNet | 12M | 76.9% | 7.6ms | +| 08 | + UDML | 12M | 78.4% | 7.6ms | +| 09 | + UIM | 12M | 79.4% | 7.6ms | -注： 测试速度时，实验01-03输入图片尺寸均为(3,32,320)，04-08输入图片尺寸均为(3,48,320)。在实际预测时，图像为变长输入，速度会有所变化。 +注： 测试速度时，实验01-03输入图片尺寸均为(3,32,320)，04-09输入图片尺寸均为(3,48,320)。在实际预测时，图像为变长输入，速度会有所变化。 **（1）轻量级文本识别网络SVTR_LCNet。** @@ -153,9 +154,10 @@ PP-OCRv3将base模型从CRNN替换成了[SVTR](https://arxiv.org/abs/2205.00159) | 02 | SVTR_Tiny | 21M | 80.1% | 97ms | | 03 | SVTR_LCNet(G4) | 9.2M | 76% | 30ms | | 04 | SVTR_LCNet(G2) | 13M | 72.98% | 9.37ms | -| 05 | SVTR_LCNet | 12M | 71.9% | 6.6ms | +| 05 | SVTR_LCNet(h32) | 12M | 71.9% | 6.6ms | +| 06 | SVTR_LCNet(h48) | 12M | 73.98% | 7.6ms | -注： 测试速度时，输入图片尺寸均为(3,32,320)； PP-OCRv2-baseline 代表没有借助蒸馏方法训练得到的模型 +注： 测试速度时，01-05输入图片尺寸均为(3,32,320)； PP-OCRv2-baseline 代表没有借助蒸馏方法训练得到的模型 **（2）采用Attention指导CTC训练。** @@ -178,7 +180,7 @@ PP-OCRv3将base模型从CRNN替换成了[SVTR](https://arxiv.org/abs/2205.00159) 为了充分利用自然场景中的大量无标注文本数据，PP-OCRv3参考论文[STR-Fewer-Labels](https://github.com/ku21fan/STR-Fewer-Labels)，设计TextRotNet自监督任务，对识别图像进行旋转并预测其旋转角度，同时结合中文场景文字识别任务的特点，在训练时适当调整图像的尺寸，添加文本识别数据增广，最终产出针对文本识别任务的PP-LCNet预训练模型，帮助识别模型精度进一步提升0.6%。TextRotNet训练流程如下图所示：
- +
@@ -187,7 +189,7 @@ PP-OCRv3将base模型从CRNN替换成了[SVTR](https://arxiv.org/abs/2205.00159) 为更直接利用自然场景中包含大量无标注数据，使用PP-OCRv2检测模型以及SVTR_tiny识别模型对百度开源的40W [LSVT弱标注数据集](https://ai.baidu.com/broad/introduction?dataset=lsvt)进行检测与识别，并筛选出识别得分大于0.95的文本，共81W文本行数据，将其补充到训练数据中，最终进一步提升模型精度1.0%。
- +
diff --git a/doc/doc_ch/models_list.md b/doc/doc_ch/models_list.md index 898133dd1d5c29870840705ceeb4561b9733d219..2012381af5a1cfe53771903e0ab99bab0b7cbc08 100644 --- a/doc/doc_ch/models_list.md +++ b/doc/doc_ch/models_list.md @@ -18,13 +18,13 @@ - [3. 文本方向分类模型](#3-文本方向分类模型) - [4. Paddle-Lite 模型](#4-paddle-lite-模型) -PaddleOCR提供的可下载模型包括`推理模型`、`训练模型`、`预训练模型`、`slim模型`，模型区别说明如下： +PaddleOCR提供的可下载模型包括`推理模型`、`训练模型`、`预训练模型`、`nb模型`，模型区别说明如下： |模型类型|模型格式|简介| |--- | --- | --- | |推理模型|inference.pdmodel、inference.pdiparams|用于预测引擎推理，[详情](./inference.md)| |训练模型、预训练模型|\*.pdparams、\*.pdopt、\*.states |训练过程中保存的模型的参数、优化器状态和训练中间信息，多用于模型指标评估和恢复训练| -|slim模型|\*.nb|经过飞桨模型压缩工具PaddleSlim压缩后的模型，适用于移动端/IoT端等端侧部署场景（需使用飞桨Paddle Lite部署）。| +|nb模型|\*.nb|经过飞桨Paddle-Lite工具优化后的模型，适用于移动端/IoT端等端侧部署场景（需使用飞桨Paddle Lite部署）。| 各个模型的关系如下面的示意图所示。 @@ -41,7 +41,7 @@ PaddleOCR提供的可下载模型包括`推理模型`、`训练模型`、`预训 |模型名称|模型简介|配置文件|推理模型大小|下载地址| | --- | --- | --- | --- | --- | -|ch_PP-OCRv3_det_slim|【最新】slim量化+蒸馏版超轻量模型，支持中英文、多语种文本检测|[ch_PP-OCRv3_det_cml.yml](../../configs/det/ch_PP-OCRv3/ch_PP-OCRv3_det_cml.yml)| 1.1M |[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_det_slim_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_det_slim_distill_train.tar) / [slim模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_det_slim_infer.nb)| +|ch_PP-OCRv3_det_slim|【最新】slim量化+蒸馏版超轻量模型，支持中英文、多语种文本检测|[ch_PP-OCRv3_det_cml.yml](../../configs/det/ch_PP-OCRv3/ch_PP-OCRv3_det_cml.yml)| 1.1M |[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_det_slim_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_det_slim_distill_train.tar) / [nb模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_det_slim_infer.nb)| |ch_PP-OCRv3_det| 【最新】原始超轻量模型，支持中英文、多语种文本检测 |[ch_PP-OCRv3_det_cml.yml](../../configs/det/ch_PP-OCRv3/ch_PP-OCRv3_det_cml.yml)| 3.8M |[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_det_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_det_distill_train.tar)| |ch_PP-OCRv2_det_slim| slim量化+蒸馏版超轻量模型，支持中英文、多语种文本检测|[ch_PP-OCRv2_det_cml.yml](../../configs/det/ch_PP-OCRv2/ch_PP-OCRv2_det_cml.yml)| 3M |[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_det_slim_quant_infer.tar)| |ch_PP-OCRv2_det| 原始超轻量模型，支持中英文、多语种文本检测|[ch_PP-OCRv2_det_cml.yml](../../configs/det/ch_PP-OCRv2/ch_PP-OCRv2_det_cml.yml)|3M|[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_det_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_det_distill_train.tar)| @@ -55,8 +55,8 @@ PaddleOCR提供的可下载模型包括`推理模型`、`训练模型`、`预训 |模型名称|模型简介|配置文件|推理模型大小|下载地址| | --- | --- | --- | --- | --- | -|en_PP-OCRv3_det_slim |【最新】slim量化版超轻量模型，支持英文、数字检测 | [ch_PP-OCRv3_det_cml.yml](../../configs/det/ch_PP-OCRv3/ch_PP-OCRv3_det_cml.yml) | 1.1M |[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_det_slim_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_det_slim_distill_train.tar) / [slim模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_det_slim_infer.nb) | -|ch_PP-OCRv3_det |【最新】原始超轻量模型，支持英文、数字检测|[ch_PP-OCRv3_det_cml.yml](../../configs/det/ch_PP-OCRv3/ch_PP-OCRv3_det_cml.yml)| 3.8M | [推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_det_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_det_distill_train.tar) | +|en_PP-OCRv3_det_slim |【最新】slim量化版超轻量模型，支持英文、数字检测 | [ch_PP-OCRv3_det_cml.yml](../../configs/det/ch_PP-OCRv3/ch_PP-OCRv3_det_cml.yml) | 1.1M |[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_det_slim_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_det_slim_distill_train.tar) / [nb模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_det_slim_infer.nb) | +|en_PP-OCRv3_det |【最新】原始超轻量模型，支持英文、数字检测|[ch_PP-OCRv3_det_cml.yml](../../configs/det/ch_PP-OCRv3/ch_PP-OCRv3_det_cml.yml)| 3.8M | [推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_det_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_det_distill_train.tar) | * 注：英文检测模型与中文检测模型结构完全相同，只有训练数据不同，在此仅提供相同的配置文件。 @@ -66,7 +66,7 @@ PaddleOCR提供的可下载模型包括`推理模型`、`训练模型`、`预训 |模型名称|模型简介|配置文件|推理模型大小|下载地址| | --- | --- | --- | --- | --- | -| ml_PP-OCRv3_det_slim |【最新】slim量化版超轻量模型，支持多语言检测 | [ch_PP-OCRv3_det_cml.yml](../../configs/det/ch_PP-OCRv3/ch_PP-OCRv3_det_cml.yml) | 1.1M |[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/Multilingual_PP-OCRv3_det_slim_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/Multilingual_PP-OCRv3_det_slim_distill_train.tar) / [slim模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/Multilingual_PP-OCRv3_det_slim_infer.nb) | +| ml_PP-OCRv3_det_slim |【最新】slim量化版超轻量模型，支持多语言检测 | [ch_PP-OCRv3_det_cml.yml](../../configs/det/ch_PP-OCRv3/ch_PP-OCRv3_det_cml.yml) | 1.1M |[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/Multilingual_PP-OCRv3_det_slim_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/Multilingual_PP-OCRv3_det_slim_distill_train.tar) / [nb模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/Multilingual_PP-OCRv3_det_slim_infer.nb) | | ml_PP-OCRv3_det |【最新】原始超轻量模型，支持多语言检测 | [ch_PP-OCRv3_det_cml.yml](../../configs/det/ch_PP-OCRv3/ch_PP-OCRv3_det_cml.yml)| 3.8M | [推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/Multilingual_PP-OCRv3_det_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/Multilingual_PP-OCRv3_det_distill_train.tar) | * 注：多语言检测模型与中文检测模型结构完全相同，只有训练数据不同，在此仅提供相同的配置文件。 @@ -81,7 +81,7 @@ PaddleOCR提供的可下载模型包括`推理模型`、`训练模型`、`预训 |模型名称|模型简介|配置文件|推理模型大小|下载地址| | --- | --- | --- | --- | --- | -|ch_PP-OCRv3_rec_slim |【最新】slim量化版超轻量模型，支持中英文、数字识别|[ch_PP-OCRv3_rec_distillation.yml](../../configs/rec/PP-OCRv3/ch_PP-OCRv3_rec_distillation.yml)| 4.9M |[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_rec_slim_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_rec_slim_train.tar) / [slim模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_rec_slim_infer.nb) | +|ch_PP-OCRv3_rec_slim |【最新】slim量化版超轻量模型，支持中英文、数字识别|[ch_PP-OCRv3_rec_distillation.yml](../../configs/rec/PP-OCRv3/ch_PP-OCRv3_rec_distillation.yml)| 4.9M |[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_rec_slim_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_rec_slim_train.tar) / [nb模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_rec_slim_infer.nb) | |ch_PP-OCRv3_rec|【最新】原始超轻量模型，支持中英文、数字识别|[ch_PP-OCRv3_rec_distillation.yml](../../configs/rec/PP-OCRv3/ch_PP-OCRv3_rec_distillation.yml)| 12.4M |[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_rec_train.tar) | |ch_PP-OCRv2_rec_slim| slim量化版超轻量模型，支持中英文、数字识别|[ch_PP-OCRv2_rec.yml](../../configs/rec/ch_PP-OCRv2/ch_PP-OCRv2_rec.yml)| 9M |[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_slim_quant_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_slim_quant_train.tar) | |ch_PP-OCRv2_rec| 原始超轻量模型，支持中英文、数字识别|[ch_PP-OCRv2_rec_distillation.yml](../../configs/rec/ch_PP-OCRv2/ch_PP-OCRv2_rec_distillation.yml)|8.5M|[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_train.tar) | @@ -96,8 +96,8 @@ PaddleOCR提供的可下载模型包括`推理模型`、`训练模型`、`预训 |模型名称|模型简介|配置文件|推理模型大小|下载地址| | --- | --- | --- | --- | --- | -|en_PP-OCRv3_rec_slim |【最新】slim量化版超轻量模型，支持英文、数字识别 | [en_PP-OCRv3_rec.yml](../../configs/rec/PP-OCRv3/en_PP-OCRv3_rec.yml)| - |[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_rec_slim_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_rec_slim_train.tar) / [slim模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_rec_slim_infer.nb) | -|ch_PP-OCRv3_rec |【最新】原始超轻量模型，支持英文、数字识别|[en_PP-OCRv3_rec.yml](../../configs/rec/en_PP-OCRv3/en_PP-OCRv3_rec.yml)| 9.6M | [推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_rec_train.tar) | +|en_PP-OCRv3_rec_slim |【最新】slim量化版超轻量模型，支持英文、数字识别 | [en_PP-OCRv3_rec.yml](../../configs/rec/PP-OCRv3/en_PP-OCRv3_rec.yml)| 3.2M |[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_rec_slim_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_rec_slim_train.tar) / [nb模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_rec_slim_infer.nb) | +|en_PP-OCRv3_rec |【最新】原始超轻量模型，支持英文、数字识别|[en_PP-OCRv3_rec.yml](../../configs/rec/en_PP-OCRv3/en_PP-OCRv3_rec.yml)| 9.6M | [推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_rec_train.tar) | |en_number_mobile_slim_v2.0_rec|slim裁剪量化版超轻量模型，支持英文、数字识别|[rec_en_number_lite_train.yml](../../configs/rec/multi_language/rec_en_number_lite_train.yml)| 2.7M | [推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/en_number_mobile_v2.0_rec_slim_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/en_number_mobile_v2.0_rec_slim_train.tar) | |en_number_mobile_v2.0_rec|原始超轻量模型，支持英文、数字识别|[rec_en_number_lite_train.yml](../../configs/rec/multi_language/rec_en_number_lite_train.yml)|2.6M|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/en_number_mobile_v2.0_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/en_number_mobile_v2.0_rec_train.tar) | @@ -107,18 +107,16 @@ PaddleOCR提供的可下载模型包括`推理模型`、`训练模型`、`预训 |模型名称|字典文件|模型简介|配置文件|推理模型大小|下载地址| | --- | --- | --- | --- |--- | --- | -| korean_PP-OCRv3_rec | ppocr/utils/dict/korean_dict.txt |韩文识别|[korean_PP-OCRv3_rec.yml](../../configs/rec/PP-OCRv3/multi_language/korean_PP-OCRv3_rec.yml)|11M|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/korean_PP-OCRv3_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/korean_PP-OCRv3_rec_train.tar) | -| japan_PP-OCRv3_rec | ppocr/utils/dict/japan_dict.txt |日文识别|[japan_PP-OCRv3_rec.yml](../../configs/rec/PP-OCRv3/multi_language/japan_PP-OCRv3_rec.yml)|11M|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/japan_PP-OCRv3_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/japan_PP-OCRv3_rec_train.tar) | -| chinese_cht_PP-OCRv3_rec | ppocr/utils/dict/chinese_cht_dict.txt | 中文繁体识别|[chinese_cht_PP-OCRv3_rec.yml](../../configs/rec/PP-OCRv3/multi_language/chinese_cht_PP-OCRv3_rec.yml)|12M|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/chinese_cht_PP-OCRv3_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/chinese_cht_PP-OCRv3_rec_train.tar) | -| te_PP-OCRv3_rec | ppocr/utils/dict/te_dict.txt | 泰卢固文识别|[te_PP-OCRv3_rec.yml](../../configs/rec/PP-OCRv3/multi_language/te_PP-OCRv3_rec.yml)|9.6M|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/te_PP-OCRv3_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/te_PP-OCRv3_rec_train.tar) | -| ka_PP-OCRv3_rec | ppocr/utils/dict/ka_dict.txt |卡纳达文识别|[ka_PP-OCRv3_rec.yml](../../configs/rec/PP-OCRv3/multi_language/ka_PP-OCRv3_rec.yml)|9.9M|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/ka_PP-OCRv3_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/ka_PP-OCRv3_rec_train.tar) | -| ta_PP-OCRv3_rec | ppocr/utils/dict/ta_dict.txt |泰米尔文识别|[ta_PP-OCRv3_rec.yml](../../configs/rec/PP-OCRv3/multi_language/ta_PP-OCRv3_rec.yml)|9.6M|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/ta_PP-OCRv3_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/ta_PP-OCRv3_rec_train.tar) | -| latin_PP-OCRv3_rec | ppocr/utils/dict/latin_dict.txt | 拉丁文识别 | [latin_PP-OCRv3_rec.yml](../../configs/rec/PP-OCRv3/multi_language/latin_PP-OCRv3_rec.yml) |9.6M|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/latin_PP-OCRv3_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/latin_PP-OCRv3_rec_train.tar) | -| arabic_PP-OCRv3_rec | ppocr/utils/dict/arabic_dict.txt | 阿拉伯字母 | [arabic_PP-OCRv3_rec.yml](../../configs/rec/PP-OCRv3/multi_language/rec_arabic_lite_train.yml) |9.6M|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/arabic_PP-OCRv3_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/arabic_PP-OCRv3_rec_train.tar) | -| cyrillic_PP-OCRv3_rec | ppocr/utils/dict/cyrillic_dict.txt | 斯拉夫字母 | [cyrillic_PP-OCRv3_rec.yml](../../configs/rec/PP-OCRv3/multi_language/cyrillic_PP-OCRv3_rec.yml) |9.6M|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/cyrillic_PP-OCRv3_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/cyrillic_PP-OCRv3_rec_train.tar) | -| devanagari_PP-OCRv3_rec | ppocr/utils/dict/devanagari_dict.txt |梵文字母 | [devanagari_PP-OCRv3_rec.yml](../../configs/rec/PP-OCRv3/multi_language/devanagari_PP-OCRv3_rec.yml) |9.6M|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/devanagari_PP-OCRv3_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/devanagari_PP-OCRv3_rec_train.tar) | - - +| korean_PP-OCRv3_rec | ppocr/utils/dict/korean_dict.txt |韩文识别|[korean_PP-OCRv3_rec.yml](../../configs/rec/PP-OCRv3/multi_language/korean_PP-OCRv3_rec.yml)|11M|[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/korean_PP-OCRv3_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/korean_PP-OCRv3_rec_train.tar) | +| japan_PP-OCRv3_rec | ppocr/utils/dict/japan_dict.txt |日文识别|[japan_PP-OCRv3_rec.yml](../../configs/rec/PP-OCRv3/multi_language/japan_PP-OCRv3_rec.yml)|11M|[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/japan_PP-OCRv3_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/japan_PP-OCRv3_rec_train.tar) | +| chinese_cht_PP-OCRv3_rec | ppocr/utils/dict/chinese_cht_dict.txt | 中文繁体识别|[chinese_cht_PP-OCRv3_rec.yml](../../configs/rec/PP-OCRv3/multi_language/chinese_cht_PP-OCRv3_rec.yml)|12M|[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/chinese_cht_PP-OCRv3_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/chinese_cht_PP-OCRv3_rec_train.tar) | +| te_PP-OCRv3_rec | ppocr/utils/dict/te_dict.txt | 泰卢固文识别|[te_PP-OCRv3_rec.yml](../../configs/rec/PP-OCRv3/multi_language/te_PP-OCRv3_rec.yml)|9.6M|[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/te_PP-OCRv3_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/te_PP-OCRv3_rec_train.tar) | +| ka_PP-OCRv3_rec | ppocr/utils/dict/ka_dict.txt |卡纳达文识别|[ka_PP-OCRv3_rec.yml](../../configs/rec/PP-OCRv3/multi_language/ka_PP-OCRv3_rec.yml)|9.9M|[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/ka_PP-OCRv3_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/ka_PP-OCRv3_rec_train.tar) | +| ta_PP-OCRv3_rec | ppocr/utils/dict/ta_dict.txt |泰米尔文识别|[ta_PP-OCRv3_rec.yml](../../configs/rec/PP-OCRv3/multi_language/ta_PP-OCRv3_rec.yml)|9.6M|[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/ta_PP-OCRv3_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/ta_PP-OCRv3_rec_train.tar) | +| latin_PP-OCRv3_rec | ppocr/utils/dict/latin_dict.txt | 拉丁文识别 | [latin_PP-OCRv3_rec.yml](../../configs/rec/PP-OCRv3/multi_language/latin_PP-OCRv3_rec.yml) |9.7M|[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/latin_PP-OCRv3_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/latin_PP-OCRv3_rec_train.tar) | +| arabic_PP-OCRv3_rec | ppocr/utils/dict/arabic_dict.txt | 阿拉伯字母 | [arabic_PP-OCRv3_rec.yml](../../configs/rec/PP-OCRv3/multi_language/rec_arabic_lite_train.yml) |9.6M|[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/arabic_PP-OCRv3_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/arabic_PP-OCRv3_rec_train.tar) | +| cyrillic_PP-OCRv3_rec | ppocr/utils/dict/cyrillic_dict.txt | 斯拉夫字母 | [cyrillic_PP-OCRv3_rec.yml](../../configs/rec/PP-OCRv3/multi_language/cyrillic_PP-OCRv3_rec.yml) |9.6M|[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/cyrillic_PP-OCRv3_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/cyrillic_PP-OCRv3_rec_train.tar) | +| devanagari_PP-OCRv3_rec | ppocr/utils/dict/devanagari_dict.txt |梵文字母 | [devanagari_PP-OCRv3_rec.yml](../../configs/rec/PP-OCRv3/multi_language/devanagari_PP-OCRv3_rec.yml) |9.9M|[推理模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/devanagari_PP-OCRv3_rec_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/devanagari_PP-OCRv3_rec_train.tar) | 更多支持语种请参考: [多语言模型](./multi_languages.md) @@ -128,13 +126,18 @@ PaddleOCR提供的可下载模型包括`推理模型`、`训练模型`、`预训 |模型名称|模型简介|配置文件|推理模型大小|下载地址| | --- | --- | --- | --- | --- | -|ch_ppocr_mobile_slim_v2.0_cls|slim量化版模型，对检测到的文本行文字角度分类|[cls_mv3.yml](../../configs/cls/cls_mv3.yml)| 2.1M |[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_slim_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_slim_infer.tar) | +|ch_ppocr_mobile_slim_v2.0_cls|slim量化版模型，对检测到的文本行文字角度分类|[cls_mv3.yml](../../configs/cls/cls_mv3.yml)| 2.1M |[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_slim_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_rec_slim_infer.tar) / [nb模型](https://paddleocr.bj.bcebos.com/PP-OCRv2/lite/ch_ppocr_mobile_v2.0_cls_infer_opt.nb) | |ch_ppocr_mobile_v2.0_cls|原始分类器模型，对检测到的文本行文字角度分类|[cls_mv3.yml](../../configs/cls/cls_mv3.yml)|1.38M|[推理模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar) / [训练模型](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_train.tar) | ## 4. Paddle-Lite 模型 +Paddle-Lite 是一个高性能、轻量级、灵活性强且易于扩展的深度学习推理框架，它可以对inference模型进一步优化，得到适用于移动端/IoT端等端侧部署场景的`nb模型`。一般建议基于量化模型进行转换，因为可以将模型以INT8形式进行存储与推理，从而进一步减小模型大小，提升模型速度。 + +本节主要列出PP-OCRv2以及更早版本的检测与识别nb模型，最新版本的nb模型可以直接从上面的模型列表中获得。 + + |模型版本|模型简介|模型大小|检测模型|文本方向分类模型|识别模型|Paddle-Lite版本| |---|---|---|---|---|---|---| |PP-OCRv2|蒸馏版超轻量中文OCR移动端模型|11M|[下载地址](https://paddleocr.bj.bcebos.com/PP-OCRv2/lite/ch_PP-OCRv2_det_infer_opt.nb)|[下载地址](https://paddleocr.bj.bcebos.com/PP-OCRv2/lite/ch_ppocr_mobile_v2.0_cls_infer_opt.nb)|[下载地址](https://paddleocr.bj.bcebos.com/PP-OCRv2/lite/ch_PP-OCRv2_rec_infer_opt.nb)|v2.10| diff --git a/doc/doc_ch/quickstart.md b/doc/doc_ch/quickstart.md index bc8e5a0a1eb0b0a299c94272142dbf21a5c75d91..29ca48fa838be4a60f08d31d5031180b951e33bc 100644 --- a/doc/doc_ch/quickstart.md +++ b/doc/doc_ch/quickstart.md @@ -59,15 +59,13 @@ cd /path/to/ppocr_img 如果不使用提供的测试图片，可以将下方`--image_dir`参数替换为相应的测试图片路径。 -**注意** whl包默认使用`PP-OCRv3`模型，识别模型使用的输入shape为`3,48,320`, 因此如果使用识别功能，需要添加参数`--rec_image_shape 3,48,320`，如果不使用默认的`PP-OCRv3`模型，则无需设置该参数。 - #### 2.1.1 中英文模型 * 检测+方向分类器+识别全流程：`--use_angle_cls true`设置使用方向分类器识别180度旋转文字，`--use_gpu false`设置不使用GPU ```bash - paddleocr --image_dir ./imgs/11.jpg --use_angle_cls true --use_gpu false --rec_image_shape 3,48,320 + paddleocr --image_dir ./imgs/11.jpg --use_angle_cls true --use_gpu false ``` 结果是一个list，每个item包含了文本框，文字和识别置信度 @@ -94,7 +92,7 @@ cd /path/to/ppocr_img - 单独使用识别：设置`--det`为`false` ```bash - paddleocr --image_dir ./imgs_words/ch/word_1.jpg --det false --rec_image_shape 3,48,320 + paddleocr --image_dir ./imgs_words/ch/word_1.jpg --det false ``` 结果是一个list，每个item只包含识别结果和识别置信度 @@ -104,16 +102,16 @@ cd /path/to/ppocr_img ``` -如需使用2.0模型，请指定参数`--version PP-OCR`，paddleocr默认使用PP-OCRv3模型(`--versioin PP-OCRv3`)。更多whl包使用可参考[whl包文档](./whl.md) +如需使用2.0模型，请指定参数`--ocr_version PP-OCR`，paddleocr默认使用PP-OCRv3模型(`--ocr_version PP-OCRv3`)。更多whl包使用可参考[whl包文档](./whl.md) #### 2.1.2 多语言模型 -PaddleOCR目前支持80个语种，可以通过修改`--lang`参数进行切换，对于英文模型，指定`--lang=en`, PP-OCRv3目前只支持中文和英文模型，其他多语言模型会陆续更新。 +PaddleOCR目前支持80个语种，可以通过修改`--lang`参数进行切换，对于英文模型，指定`--lang=en`。 ``` bash -paddleocr --image_dir ./imgs_en/254.jpg --lang=en --rec_image_shape 3,48,320 +paddleocr --image_dir ./imgs_en/254.jpg --lang=en ```
diff --git a/doc/doc_ch/update.md b/doc/doc_ch/update.md index 9071e673910f8d87762dc8f9dd097d444f36e624..69dde041d3b4d6af3b248b9392a7946dca61858e 100644 --- a/doc/doc_ch/update.md +++ b/doc/doc_ch/update.md @@ -1,4 +1,5 @@ # 更新 +- 2022.5.7 添加对[Weights & Biases](https://docs.wandb.ai/)训练日志记录工具的支持。 - 2021.12.21 《OCR十讲》课程开讲，12月21日起每晚八点半线上授课！ 【免费】报名地址：https://aistudio.baidu.com/aistudio/course/introduce/25207 - 2021.12.21 发布PaddleOCR v2.4。OCR算法新增1种文本检测算法（PSENet），3种文本识别算法（NRTR、SEED、SAR）；文档结构化算法新增1种关键信息提取算法（SDMGR），3种DocVQA算法（LayoutLM、LayoutLMv2，LayoutXLM）。 - 2021.9.7 发布PaddleOCR v2.3，发布[PP-OCRv2](#PP-OCRv2)，CPU推理速度相比于PP-OCR server提升220%；效果相比于PP-OCR mobile 提升7%。 diff --git a/doc/doc_ch/whl.md b/doc/doc_ch/whl.md index d57f2ac3255a78b630e2ea4189ab182d5c7f71ba..511e0421f1e249e340f2002a900b59633e31880e 100644 --- a/doc/doc_ch/whl.md +++ b/doc/doc_ch/whl.md @@ -199,12 +199,10 @@ for line in result: paddleocr -h ``` -**注意** whl包默认使用`PP-OCRv3`模型，识别模型使用的输入shape为`3,48,320`, 因此如果使用识别功能，需要添加参数`--rec_image_shape 3,48,320`，如果不使用默认的`PP-OCRv3`模型，则无需设置该参数。 - * 检测+方向分类器+识别全流程 ```bash -paddleocr --image_dir PaddleOCR/doc/imgs/11.jpg --use_angle_cls true --rec_image_shape 3,48,320 +paddleocr --image_dir PaddleOCR/doc/imgs/11.jpg --use_angle_cls true ``` 结果是一个list，每个item包含了文本框，文字和识别置信度 @@ -217,7 +215,7 @@ paddleocr --image_dir PaddleOCR/doc/imgs/11.jpg --use_angle_cls true --rec_image * 检测+识别 ```bash -paddleocr --image_dir PaddleOCR/doc/imgs/11.jpg --rec_image_shape 3,48,320 +paddleocr --image_dir PaddleOCR/doc/imgs/11.jpg ``` 结果是一个list，每个item包含了文本框，文字和识别置信度 @@ -230,7 +228,7 @@ paddleocr --image_dir PaddleOCR/doc/imgs/11.jpg --rec_image_shape 3,48,320 * 方向分类器+识别 ```bash -paddleocr --image_dir PaddleOCR/doc/imgs_words/ch/word_1.jpg --use_angle_cls true --det false --rec_image_shape 3,48,320 +paddleocr --image_dir PaddleOCR/doc/imgs_words/ch/word_1.jpg --use_angle_cls true --det false ``` 结果是一个list，每个item只包含识别结果和识别置信度 @@ -256,7 +254,7 @@ paddleocr --image_dir PaddleOCR/doc/imgs/11.jpg --rec false * 单独执行识别 ```bash -paddleocr --image_dir PaddleOCR/doc/imgs_words/ch/word_1.jpg --det false --rec_image_shape 3,48,320 +paddleocr --image_dir PaddleOCR/doc/imgs_words/ch/word_1.jpg --det false ``` 结果是一个list，每个item只包含识别结果和识别置信度 @@ -416,4 +414,4 @@ im_show.save('result.jpg') | cls | 前向时是否启动分类 (命令行模式下使用use_angle_cls控制前向是否启动分类) | FALSE | | show_log | 是否打印logger信息 | FALSE | | type | 执行ocr或者表格结构化, 值可选['ocr','structure'] | ocr | -| ocr_version | OCR模型版本，可选PP-OCRv3, PP-OCRv2, PP-OCR。PP-OCRv3 目前仅支持中、英文的检测和识别模型，方向分类器模型；PP-OCRv2 目前仅支持中文的检测和识别模型；PP-OCR支持中文的检测，识别，多语种识别，方向分类器等模型 | PP-OCRv3 | +| ocr_version | OCR模型版本，可选PP-OCRv3, PP-OCRv2, PP-OCR。PP-OCRv3 支持中、英文的检测、识别、多语种识别，方向分类器等模型；PP-OCRv2 目前仅支持中文的检测和识别模型；PP-OCR支持中文的检测，识别，多语种识别，方向分类器等模型 | PP-OCRv3 | diff --git a/doc/doc_en/models_list_en.md b/doc/doc_en/models_list_en.md index 6e61e8d0a78d8dc1e4312fe11f5e9feee88a483d..15a7fdb94e303297f7be681f297a5e52613a268a 100644 --- a/doc/doc_en/models_list_en.md +++ b/doc/doc_en/models_list_en.md @@ -16,13 +16,13 @@ - [3. Text Angle Classification Model](#3-text-angle-classification-model) - [4. Paddle-Lite Model](#4-paddle-lite-model) -The downloadable models provided by PaddleOCR include `inference model`, `trained model`, `pre-trained model` and `slim model`. The differences between the models are as follows: +The downloadable models provided by PaddleOCR include `inference model`, `trained model`, `pre-trained model` and `nb model`. The differences between the models are as follows: |model type|model format|description| |--- | --- | --- | |inference model|inference.pdmodel、inference.pdiparams|Used for inference based on Paddle inference engine，[detail](./inference_en.md)| |trained model, pre-trained model|\*.pdparams、\*.pdopt、\*.states |The checkpoints model saved in the training process, which stores the parameters of the model, mostly used for model evaluation and continuous training.| -|slim model|\*.nb| Model compressed by PaddleSlim (a model compression tool using PaddlePaddle), which is suitable for mobile-side deployment scenarios (Paddle-Lite is needed for slim model deployment). | +|nb model|\*.nb| Model optimized by Paddle-Lite, which is suitable for mobile-side deployment scenarios (Paddle-Lite is needed for nb model deployment). | Relationship of the above models is as follows. @@ -37,7 +37,7 @@ Relationship of the above models is as follows. |model name|description|config|model size|download| | --- | --- | --- | --- | --- | -|ch_PP-OCRv3_det_slim| [New] slim quantization with distillation lightweight model, supporting Chinese, English, multilingual text detection |[ch_PP-OCRv3_det_cml.yml](../../configs/det/ch_PP-OCRv3/ch_PP-OCRv3_det_cml.yml)| 1.1M |[inference model](https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_det_slim_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/PP-OCRv3/ch/ch_PP-OCRv3_det_slim_distill_train.tar) / [slim model](https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_det_slim_infer.nb)| +|ch_PP-OCRv3_det_slim| [New] slim quantization with distillation lightweight model, supporting Chinese, English, multilingual text detection |[ch_PP-OCRv3_det_cml.yml](../../configs/det/ch_PP-OCRv3/ch_PP-OCRv3_det_cml.yml)| 1.1M |[inference model](https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_det_slim_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/PP-OCRv3/ch/ch_PP-OCRv3_det_slim_distill_train.tar) / [nb model](https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_det_slim_infer.nb)| |ch_PP-OCRv3_det| [New] Original lightweight model, supporting Chinese, English, multilingual text detection |[ch_PP-OCRv3_det_cml.yml](../../configs/det/ch_PP-OCRv3/ch_PP-OCRv3_det_cml.yml)| 3.8M |[inference model](https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_det_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_det_distill_train.tar)| |ch_PP-OCRv2_det_slim| [New] slim quantization with distillation lightweight model, supporting Chinese, English, multilingual text detection|[ch_PP-OCRv2_det_cml.yml](../../configs/det/ch_PP-OCRv2/ch_PP-OCRv2_det_cml.yml)| 3M |[inference model](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_det_slim_quant_infer.tar)| |ch_PP-OCRv2_det| [New] Original lightweight model, supporting Chinese, English, multilingual text detection|[ch_PP-OCRv2_det_cml.yml](../../configs/det/ch_PP-OCRv2/ch_PP-OCRv2_det_cml.yml)|3M|[inference model](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_det_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_det_distill_train.tar)| @@ -51,7 +51,7 @@ Relationship of the above models is as follows. |model name|description|config|model size|download| | --- | --- | --- | --- | --- | -|en_PP-OCRv3_det_slim | [New] Slim qunatization with distillation lightweight detection model, supporting English | [ch_PP-OCRv3_det_cml.yml](../../configs/det/ch_PP-OCRv3/ch_PP-OCRv3_det_cml.yml) | 1.1M |[inference model](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_det_slim_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_det_slim_distill_train.tar) / [slim model](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_det_slim_infer.nb) | +|en_PP-OCRv3_det_slim | [New] Slim qunatization with distillation lightweight detection model, supporting English | [ch_PP-OCRv3_det_cml.yml](../../configs/det/ch_PP-OCRv3/ch_PP-OCRv3_det_cml.yml) | 1.1M |[inference model](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_det_slim_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_det_slim_distill_train.tar) / [nb model](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_det_slim_infer.nb) | |ch_PP-OCRv3_det | [New] Original lightweight detection model, supporting English |[ch_PP-OCRv3_det_cml.yml](../../configs/det/ch_PP-OCRv3/ch_PP-OCRv3_det_cml.yml)| 3.8M | [inference model](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_det_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_det_distill_train.tar) | * Note: English configuration file is same as Chinese except training data, here we only provide one configuration file. @@ -62,7 +62,7 @@ Relationship of the above models is as follows. |model name|description|config|model size|download| | --- | --- | --- | --- | --- | -| ml_PP-OCRv3_det_slim | [New] Slim qunatization with distillation lightweight detection model, supporting English | [ch_PP-OCRv3_det_cml.yml](../../configs/det/ch_PP-OCRv3/ch_PP-OCRv3_det_cml.yml) | 1.1M | [inference model](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/Multilingual_PP-OCRv3_det_slim_infer.tar) / [trained model ](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/Multilingual_PP-OCRv3_det_slim_distill_train.tar) / [slim model](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/Multilingual_PP-OCRv3_det_slim_infer.nb) | +| ml_PP-OCRv3_det_slim | [New] Slim qunatization with distillation lightweight detection model, supporting English | [ch_PP-OCRv3_det_cml.yml](../../configs/det/ch_PP-OCRv3/ch_PP-OCRv3_det_cml.yml) | 1.1M | [inference model](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/Multilingual_PP-OCRv3_det_slim_infer.tar) / [trained model ](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/Multilingual_PP-OCRv3_det_slim_distill_train.tar) / [nb model](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/Multilingual_PP-OCRv3_det_slim_infer.nb) | | ml_PP-OCRv3_det |[New] Original lightweight detection model, supporting English | [ch_PP-OCRv3_det_cml.yml](../../configs/det/ch_PP-OCRv3/ch_PP-OCRv3_det_cml.yml)| 3.8M | [inference model](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/Multilingual_PP-OCRv3_det_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/Multilingual_PP-OCRv3_det_distill_train.tar) | * Note: English configuration file is same as Chinese except training data, here we only provide one configuration file. @@ -75,7 +75,7 @@ Relationship of the above models is as follows. |model name|description|config|model size|download| | --- | --- | --- | --- | --- | -|ch_PP-OCRv3_rec_slim | [New] Slim qunatization with distillation lightweight model, supporting Chinese, English text recognition |[ch_PP-OCRv3_rec_distillation.yml](../../configs/rec/PP-OCRv3/ch_PP-OCRv3_rec_distillation.yml)| 4.9M |[inference model](https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_rec_slim_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/PP-OCRv3/ch/ch_PP-OCRv3_rec_slim_train.tar) / [slim model](https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_rec_slim_infer.nb) | +|ch_PP-OCRv3_rec_slim | [New] Slim qunatization with distillation lightweight model, supporting Chinese, English text recognition |[ch_PP-OCRv3_rec_distillation.yml](../../configs/rec/PP-OCRv3/ch_PP-OCRv3_rec_distillation.yml)| 4.9M |[inference model](https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_rec_slim_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/PP-OCRv3/ch/ch_PP-OCRv3_rec_slim_train.tar) / [nb model](https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_rec_slim_infer.nb) | |ch_PP-OCRv3_rec| [New] Original lightweight model, supporting Chinese, English, multilingual text recognition |[ch_PP-OCRv3_rec_distillation.yml](../../configs/rec/PP-OCRv3/ch_PP-OCRv3_rec_distillation.yml)| 12.4M |[inference model](https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_rec_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/PP-OCRv3/chinese/ch_PP-OCRv3_rec_train.tar) | |ch_PP-OCRv2_rec_slim| Slim qunatization with distillation lightweight model, supporting Chinese, English text recognition|[ch_PP-OCRv2_rec.yml](../../configs/rec/ch_PP-OCRv2/ch_PP-OCRv2_rec.yml)| 9M |[inference model](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_slim_quant_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_slim_quant_train.tar) | |ch_PP-OCRv2_rec| Original lightweight model, supporting Chinese, English, multilingual text recognition |[ch_PP-OCRv2_rec_distillation.yml](../../configs/rec/ch_PP-OCRv2/ch_PP-OCRv2_rec_distillation.yml)|8.5M|[inference model](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/PP-OCRv2/chinese/ch_PP-OCRv2_rec_train.tar) | @@ -91,8 +91,8 @@ Relationship of the above models is as follows. |model name|description|config|model size|download| | --- | --- | --- | --- | --- | -|en_PP-OCRv3_rec_slim | [New] Slim qunatization with distillation lightweight model, supporting english, English text recognition |[en_PP-OCRv3_rec_distillation.yml](../../configs/rec/PP-OCRv3/en_PP-OCRv3_rec_distillation.yml)| 4.9M |[inference model(coming soon)](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_rec_slim_infer.tar) / [trained model (coming soon)](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_rec_slim_train.tar) / [slim model(coming soon)](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_rec_slim_infer.nb) | -|en_PP-OCRv3_rec| [New] Original lightweight model, supporting english, English, multilingual text recognition |[en_PP-OCRv3_rec_distillation.yml](../../configs/rec/PP-OCRv3/en_PP-OCRv3_rec_distillation.yml)| 12.4M |[inference model](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_rec_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_rec_train.tar) | +|en_PP-OCRv3_rec_slim | [New] Slim qunatization with distillation lightweight model, supporting english, English text recognition |[en_PP-OCRv3_rec.yml](../../configs/rec/PP-OCRv3/en_PP-OCRv3_rec.yml)| 3.2M |[inference model](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_rec_slim_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_rec_slim_train.tar) / [nb model](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_rec_slim_infer.nb) | +|en_PP-OCRv3_rec| [New] Original lightweight model, supporting english, English, multilingual text recognition |[en_PP-OCRv3_rec.yml](../../configs/rec/PP-OCRv3/en_PP-OCRv3_rec.yml)| 9.6M |[inference model](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_rec_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_rec_train.tar) | |en_number_mobile_slim_v2.0_rec|Slim pruned and quantized lightweight model, supporting English and number recognition|[rec_en_number_lite_train.yml](../../configs/rec/multi_language/rec_en_number_lite_train.yml)| 2.7M | [inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/en_number_mobile_v2.0_rec_slim_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/en/en_number_mobile_v2.0_rec_slim_train.tar) | |en_number_mobile_v2.0_rec|Original lightweight model, supporting English and number recognition|[rec_en_number_lite_train.yml](../../configs/rec/multi_language/rec_en_number_lite_train.yml)|2.6M|[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/en_number_mobile_v2.0_rec_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/multilingual/en_number_mobile_v2.0_rec_train.tar) | @@ -122,11 +122,16 @@ For more supported languages, please refer to : [Multi-language model](./multi_l |model name|description|config|model size|download| | --- | --- | --- | --- | --- | -|ch_ppocr_mobile_slim_v2.0_cls|Slim quantized model for text angle classification|[cls_mv3.yml](../../configs/cls/cls_mv3.yml)| 2.1M | [inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_slim_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_slim_train.tar) | +|ch_ppocr_mobile_slim_v2.0_cls|Slim quantized model for text angle classification|[cls_mv3.yml](../../configs/cls/cls_mv3.yml)| 2.1M | [inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_slim_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_slim_train.tar) / [nb model](https://paddleocr.bj.bcebos.com/PP-OCRv2/lite/ch_ppocr_mobile_v2.0_cls_infer_opt.nb) | |ch_ppocr_mobile_v2.0_cls|Original model for text angle classification|[cls_mv3.yml](../../configs/cls/cls_mv3.yml)|1.38M|[inference model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_infer.tar) / [trained model](https://paddleocr.bj.bcebos.com/dygraph_v2.0/ch/ch_ppocr_mobile_v2.0_cls_train.tar) | ## 4. Paddle-Lite Model + +Paddle Lite is an updated version of Paddle-Mobile, an open-open source deep learning framework designed to make it easy to perform inference on mobile, embeded, and IoT devices. It can further optimize the inference model and generate `nb model` used for edge devices. It's suggested to optimize the quantization model using Paddle-Lite because `INT8` format is used for the model storage and inference. + +This chapter lists OCR nb models with PP-OCRv2 or earlier versions. You can access to the latest nb models from the above tables. + |Version|Introduction|Model size|Detection model|Text Direction model|Recognition model|Paddle-Lite branch| |---|---|---|---|---|---|---| |PP-OCRv2|extra-lightweight chinese OCR optimized model|11M|[download link](https://paddleocr.bj.bcebos.com/PP-OCRv2/lite/ch_PP-OCRv2_det_infer_opt.nb)|[download link](https://paddleocr.bj.bcebos.com/PP-OCRv2/lite/ch_ppocr_mobile_v2.0_cls_infer_opt.nb)|[download link](https://paddleocr.bj.bcebos.com/PP-OCRv2/lite/ch_PP-OCRv2_rec_infer_opt.nb)|v2.10| diff --git a/doc/doc_en/quickstart_en.md b/doc/doc_en/quickstart_en.md index 4a31924c7993960f1510816033b68af08c79b9a0..d7aeb7773021aa6cf8f4d71298588915e5938fab 100644 --- a/doc/doc_en/quickstart_en.md +++ b/doc/doc_en/quickstart_en.md @@ -73,8 +73,6 @@ cd /path/to/ppocr_img If you do not use the provided test image, you can replace the following `--image_dir` parameter with the corresponding test image path -**Note**: The whl package uses the `PP-OCRv3` model by default, and the input shape used by the recognition model is `3,48,320`, so if you use the recognition function, you need to add the parameter `--rec_image_shape 3,48,320`, if you do not use the default `PP- OCRv3` model, you do not need to set this parameter. - #### 2.1.1 Chinese and English Model @@ -82,7 +80,7 @@ If you do not use the provided test image, you can replace the following `--imag * Detection, direction classification and recognition: set the parameter`--use_gpu false` to disable the gpu device ```bash - paddleocr --image_dir ./imgs_en/img_12.jpg --use_angle_cls true --lang en --use_gpu false --rec_image_shape 3,48,320 + paddleocr --image_dir ./imgs_en/img_12.jpg --use_angle_cls true --lang en --use_gpu false ``` Output will be a list, each item contains bounding box, text and recognition confidence @@ -112,7 +110,7 @@ If you do not use the provided test image, you can replace the following `--imag * Only recognition: set `--det` to `false` ```bash - paddleocr --image_dir ./imgs_words_en/word_10.png --det false --lang en --rec_image_shape 3,48,320 + paddleocr --image_dir ./imgs_words_en/word_10.png --det false --lang en ``` Output will be a list, each item contains text and recognition confidence @@ -121,15 +119,15 @@ If you do not use the provided test image, you can replace the following `--imag ['PAIN', 0.9934559464454651] ``` -If you need to use the 2.0 model, please specify the parameter `--version PP-OCR`, paddleocr uses the PP-OCRv3 model by default(`--versioin PP-OCRv3`). More whl package usage can be found in [whl package](./whl_en.md) +If you need to use the 2.0 model, please specify the parameter `--ocr_version PP-OCR`, paddleocr uses the PP-OCRv3 model by default(`--ocr_version PP-OCRv3`). More whl package usage can be found in [whl package](./whl_en.md) #### 2.1.2 Multi-language Model -PaddleOCR currently supports 80 languages, which can be switched by modifying the `--lang` parameter. PP-OCRv3 currently only supports Chinese and English models, and other multilingual models will be updated one after another. +PaddleOCR currently supports 80 languages, which can be switched by modifying the `--lang` parameter. ``` bash -paddleocr --image_dir ./doc/imgs_en/254.jpg --lang=en --rec_image_shape 3,48,320 +paddleocr --image_dir ./doc/imgs_en/254.jpg --lang=en ```
@@ -210,4 +208,4 @@ Visualization of results In this section, you have mastered the use of PaddleOCR whl package. -PaddleOCR is a rich and practical OCR tool library that get through the whole process of data production, model training, compression, inference and deployment, please refer to the [tutorials](../../README.md#tutorials) to start the journey of PaddleOCR. \ No newline at end of file +PaddleOCR is a rich and practical OCR tool library that get through the whole process of data production, model training, compression, inference and deployment, please refer to the [tutorials](../../README.md#tutorials) to start the journey of PaddleOCR. diff --git a/doc/doc_en/update_en.md b/doc/doc_en/update_en.md index 8ec74fe8b73d89cc97904e2ce156e14bbd596eb4..24e342a636abc3300455c5e1c3d9f5670e0a9be4 100644 --- a/doc/doc_en/update_en.md +++ b/doc/doc_en/update_en.md @@ -1,4 +1,5 @@ # RECENT UPDATES +- 2022.5.7 Add support for metric and model logging during training to [Weights & Biases](https://docs.wandb.ai/). - 2021.12.21 OCR open source online course starts. The lesson starts at 8:30 every night and lasts for ten days. Free registration: https://aistudio.baidu.com/aistudio/course/introduce/25207 - 2021.12.21 release PaddleOCR v2.4, release 1 text detection algorithm (PSENet), 3 text recognition algorithms (NRTR、SEED、SAR), 1 key information extraction algorithm (SDMGR) and 3 DocVQA algorithms (LayoutLM、LayoutLMv2，LayoutXLM). - 2021.9.7 release PaddleOCR v2.3, [PP-OCRv2](#PP-OCRv2) is proposed. The CPU inference speed of PP-OCRv2 is 220% higher than that of PP-OCR server. The F-score of PP-OCRv2 is 7% higher than that of PP-OCR mobile. diff --git a/doc/doc_en/whl_en.md b/doc/doc_en/whl_en.md index 40a2e122d19679a59e7e65df29dd59781b4a2143..d81e5532cf1db0193abf61b972420bdc3bacfd0b 100644 --- a/doc/doc_en/whl_en.md +++ b/doc/doc_en/whl_en.md @@ -172,11 +172,9 @@ show help information paddleocr -h ``` -**Note**: The whl package uses the `PP-OCRv3` model by default, and the input shape used by the recognition model is `3,48,320`, so if you use the recognition function, you need to add the parameter `--rec_image_shape 3,48,320`, if you do not use the default `PP- OCRv3` model, you do not need to set this parameter. - * detection classification and recognition ```bash -paddleocr --image_dir PaddleOCR/doc/imgs_en/img_12.jpg --use_angle_cls true --lang en --rec_image_shape 3,48,320 +paddleocr --image_dir PaddleOCR/doc/imgs_en/img_12.jpg --use_angle_cls true --lang en ``` Output will be a list, each item contains bounding box, text and recognition confidence @@ -189,7 +187,7 @@ Output will be a list, each item contains bounding box, text and recognition con * detection and recognition ```bash -paddleocr --image_dir PaddleOCR/doc/imgs_en/img_12.jpg --lang en --rec_image_shape 3,48,320 +paddleocr --image_dir PaddleOCR/doc/imgs_en/img_12.jpg --lang en ``` Output will be a list, each item contains bounding box, text and recognition confidence @@ -202,7 +200,7 @@ Output will be a list, each item contains bounding box, text and recognition con * classification and recognition ```bash -paddleocr --image_dir PaddleOCR/doc/imgs_words_en/word_10.png --use_angle_cls true --det false --lang en --rec_image_shape 3,48,320 +paddleocr --image_dir PaddleOCR/doc/imgs_words_en/word_10.png --use_angle_cls true --det false --lang en ``` Output will be a list, each item contains text and recognition confidence @@ -225,7 +223,7 @@ Output will be a list, each item only contains bounding box * only recognition ```bash -paddleocr --image_dir PaddleOCR/doc/imgs_words_en/word_10.png --det false --lang en --rec_image_shape 3,48,320 +paddleocr --image_dir PaddleOCR/doc/imgs_words_en/word_10.png --det false --lang en ``` Output will be a list, each item contains text and recognition confidence @@ -368,4 +366,4 @@ im_show.save('result.jpg') | cls | Enable classification when `ppocr.ocr` func exec((Use use_angle_cls in command line mode to control whether to start classification in the forward direction) | FALSE | | show_log | Whether to print log| FALSE | | type | Perform ocr or table structuring, the value is selected in ['ocr','structure'] | ocr | -| ocr_version | OCR Model version number, the current model support list is as follows: PP-OCRv3 support Chinese and English detection and recognition model and direction classifier model, PP-OCRv2 support Chinese detection and recognition model, PP-OCR support Chinese detection, recognition and direction classifier, multilingual recognition model | PP-OCRv3 | +| ocr_version | OCR Model version number, the current model support list is as follows: PP-OCRv3 supports Chinese and English detection, recognition, multilingual recognition, direction classifier models, PP-OCRv2 support Chinese detection and recognition model, PP-OCR support Chinese detection, recognition and direction classifier, multilingual recognition model | PP-OCRv3 | diff --git a/doc/imgs_words/arabic/ar_1.jpg b/doc/imgs_words/arabic/ar_1.jpg index 33192651f8491be38373fabe2a8aec43fcd22a41..71d7bf252d73a6139a6129ec2bce9dad77920ce9 100644 Binary files a/doc/imgs_words/arabic/ar_1.jpg and b/doc/imgs_words/arabic/ar_1.jpg differ diff --git a/doc/imgs_words/arabic/ar_2.jpg b/doc/imgs_words/arabic/ar_2.jpg index 66c10840a090c674c143abf7296219876dd05817..017d3f6fbc3650d5e6a61ca4a25f6cb81232d8a5 100644 Binary files a/doc/imgs_words/arabic/ar_2.jpg and b/doc/imgs_words/arabic/ar_2.jpg differ diff --git a/paddleocr.py b/paddleocr.py index 417350839ac4d1e512c7396831f89ab4b2d6c724..f7871db6470c75db82e8251dff5361c099c4adda 100644 --- a/paddleocr.py +++ b/paddleocr.py @@ -67,6 +67,10 @@ MODEL_URLS = { 'url': 'https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_det_infer.tar', }, + 'ml': { + 'url': + 'https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/Multilingual_PP-OCRv3_det_infer.tar' + } }, 'rec': { 'ch': { @@ -79,6 +83,56 @@ MODEL_URLS = { 'https://paddleocr.bj.bcebos.com/PP-OCRv3/english/en_PP-OCRv3_rec_infer.tar', 'dict_path': './ppocr/utils/en_dict.txt' }, + 'korean': { + 'url': + 'https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/korean_PP-OCRv3_rec_infer.tar', + 'dict_path': './ppocr/utils/dict/korean_dict.txt' + }, + 'japan': { + 'url': + 'https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/japan_PP-OCRv3_rec_infer.tar', + 'dict_path': './ppocr/utils/dict/japan_dict.txt' + }, + 'chinese_cht': { + 'url': + 'https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/chinese_cht_PP-OCRv3_rec_infer.tar', + 'dict_path': './ppocr/utils/dict/chinese_cht_dict.txt' + }, + 'ta': { + 'url': + 'https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/ta_PP-OCRv3_rec_infer.tar', + 'dict_path': './ppocr/utils/dict/ta_dict.txt' + }, + 'te': { + 'url': + 'https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/te_PP-OCRv3_rec_infer.tar', + 'dict_path': './ppocr/utils/dict/te_dict.txt' + }, + 'ka': { + 'url': + 'https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/ka_PP-OCRv3_rec_infer.tar', + 'dict_path': './ppocr/utils/dict/ka_dict.txt' + }, + 'latin': { + 'url': + 'https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/latin_PP-OCRv3_rec_infer.tar', + 'dict_path': './ppocr/utils/dict/latin_dict.txt' + }, + 'arabic': { + 'url': + 'https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/arabic_PP-OCRv3_rec_infer.tar', + 'dict_path': './ppocr/utils/dict/arabic_dict.txt' + }, + 'cyrillic': { + 'url': + 'https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/cyrillic_PP-OCRv3_rec_infer.tar', + 'dict_path': './ppocr/utils/dict/cyrillic_dict.txt' + }, + 'devanagari': { + 'url': + 'https://paddleocr.bj.bcebos.com/PP-OCRv3/multilingual/devanagari_PP-OCRv3_rec_infer.tar', + 'dict_path': './ppocr/utils/dict/devanagari_dict.txt' + }, }, 'cls': { 'ch': { @@ -259,7 +313,7 @@ def parse_lang(lang): 'af', 'az', 'bs', 'cs', 'cy', 'da', 'de', 'es', 'et', 'fr', 'ga', 'hr', 'hu', 'id', 'is', 'it', 'ku', 'la', 'lt', 'lv', 'mi', 'ms', 'mt', 'nl', 'no', 'oc', 'pi', 'pl', 'pt', 'ro', 'rs_latin', 'sk', 'sl', 'sq', 'sv', - 'sw', 'tl', 'tr', 'uz', 'vi' + 'sw', 'tl', 'tr', 'uz', 'vi', 'french', 'german' ] arabic_lang = ['ar', 'fa', 'ug', 'ur'] cyrillic_lang = [ @@ -285,8 +339,10 @@ def parse_lang(lang): det_lang = "ch" elif lang == 'structure': det_lang = 'structure' - else: + elif lang in ["en", "latin"]: det_lang = "en" + else: + det_lang = "ml" return lang, det_lang @@ -356,6 +412,10 @@ class PaddleOCR(predict_system.TextSystem): params.cls_model_dir, cls_url = confirm_model_dir_url( params.cls_model_dir, os.path.join(BASE_DIR, 'whl', 'cls'), cls_model_config['url']) + if params.ocr_version == 'PP-OCRv3': + params.rec_image_shape = "3, 48, 320" + else: + params.rec_image_shape = "3, 32, 320" # download model maybe_download(params.det_model_dir, det_url) maybe_download(params.rec_model_dir, rec_url)