test_hubserving.py 4.6 KB
Newer Older
D
dyning 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
# Copyright (c) 2020 PaddlePaddle Authors. All Rights Reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.
import os
import sys
__dir__ = os.path.dirname(os.path.abspath(__file__))
sys.path.append(__dir__)
sys.path.append(os.path.abspath(os.path.join(__dir__, '..')))

M
MissPenguin 已提交
20 21 22
from ppocr.utils.logging import get_logger
logger = get_logger()

D
dyning 已提交
23 24 25 26 27
import cv2
import numpy as np
import time
from PIL import Image
from ppocr.utils.utility import get_image_file_list
文幕地方's avatar
文幕地方 已提交
28 29 30
from tools.infer.utility import draw_ocr, draw_boxes, str2bool
from ppstructure.utility import draw_structure_result
from ppstructure.predict_system import save_structure_res, to_excel
D
dyning 已提交
31 32 33 34

import requests
import json
import base64
D
dyning 已提交
35

D
dyning 已提交
36 37 38 39

def cv2_to_base64(image):
    return base64.b64encode(image).decode('utf8')

D
dyning 已提交
40 41 42 43 44 45 46

def draw_server_result(image_file, res):
    img = cv2.imread(image_file)
    image = Image.fromarray(cv2.cvtColor(img, cv2.COLOR_BGR2RGB))
    if len(res) == 0:
        return np.array(image)
    keys = res[0].keys()
littletomatodonkey's avatar
littletomatodonkey 已提交
47 48
    if 'text_region' not in keys:  # for ocr_rec, draw function is invalid 
        logger.info("draw function is invalid for ocr_rec!")
D
dyning 已提交
49
        return None
littletomatodonkey's avatar
littletomatodonkey 已提交
50 51
    elif 'text' not in keys:  # for ocr_det
        logger.info("draw text boxes only!")
D
dyning 已提交
52 53 54 55 56 57
        boxes = []
        for dno in range(len(res)):
            boxes.append(res[dno]['text_region'])
        boxes = np.array(boxes)
        draw_img = draw_boxes(image, boxes)
        return draw_img
littletomatodonkey's avatar
littletomatodonkey 已提交
58 59
    else:  # for ocr_system
        logger.info("draw boxes and texts!")
D
dyning 已提交
60 61 62 63 64 65 66 67 68
        boxes = []
        texts = []
        scores = []
        for dno in range(len(res)):
            boxes.append(res[dno]['text_region'])
            texts.append(res[dno]['text'])
            scores.append(res[dno]['confidence'])
        boxes = np.array(boxes)
        scores = np.array(scores)
littletomatodonkey's avatar
littletomatodonkey 已提交
69 70
        draw_img = draw_ocr(
            image, boxes, texts, scores, draw_txt=True, drop_score=0.5)
D
dyning 已提交
71 72 73
        return draw_img


文幕地方's avatar
文幕地方 已提交
74 75
def main(args):
    image_file_list = get_image_file_list(args.image_dir)
D
dyning 已提交
76 77 78 79 80 81 82 83 84
    is_visualize = False
    headers = {"Content-type": "application/json"}
    cnt = 0
    total_time = 0
    for image_file in image_file_list:
        img = open(image_file, 'rb').read()
        if img is None:
            logger.info("error in loading image:{}".format(image_file))
            continue
文幕地方's avatar
文幕地方 已提交
85
        img_name = os.path.basename(image_file)
D
dyning 已提交
86 87
        # 发送HTTP请求
        starttime = time.time()
littletomatodonkey's avatar
littletomatodonkey 已提交
88
        data = {'images': [cv2_to_base64(img)]}
文幕地方's avatar
文幕地方 已提交
89 90
        r = requests.post(
            url=args.server_url, headers=headers, data=json.dumps(data))
D
dyning 已提交
91 92
        elapse = time.time() - starttime
        total_time += elapse
littletomatodonkey's avatar
littletomatodonkey 已提交
93
        logger.info("Predict time of %s: %.3fs" % (image_file, elapse))
D
dyning 已提交
94
        res = r.json()["results"][0]
littletomatodonkey's avatar
littletomatodonkey 已提交
95
        logger.info(res)
D
dyning 已提交
96

文幕地方's avatar
文幕地方 已提交
97 98 99 100 101 102 103 104
        if args.visualize:
            draw_img = None
            if 'structure_table' in args.server_url:
                to_excel(res, './{}.xlsx'.format(img_name))
            elif 'structure_system' in args.server_url:
                pass
            else:
                draw_img = draw_server_result(image_file, res)
D
dyning 已提交
105 106 107 108 109 110 111
            if draw_img is not None:
                draw_img_save = "./server_results/"
                if not os.path.exists(draw_img_save):
                    os.makedirs(draw_img_save)
                cv2.imwrite(
                    os.path.join(draw_img_save, os.path.basename(image_file)),
                    draw_img[:, :, ::-1])
littletomatodonkey's avatar
littletomatodonkey 已提交
112
                logger.info("The visualized image saved in {}".format(
D
dyning 已提交
113 114 115
                    os.path.join(draw_img_save, os.path.basename(image_file))))
        cnt += 1
        if cnt % 100 == 0:
littletomatodonkey's avatar
littletomatodonkey 已提交
116 117
            logger.info("{} processed".format(cnt))
    logger.info("avg time cost: {}".format(float(total_time) / cnt))
D
dyning 已提交
118

littletomatodonkey's avatar
littletomatodonkey 已提交
119

文幕地方's avatar
文幕地方 已提交
120 121 122 123 124 125 126 127 128 129
def parse_args():
    import argparse
    parser = argparse.ArgumentParser(description="args for hub serving")
    parser.add_argument("--server_url", type=str, required=True)
    parser.add_argument("--image_dir", type=str, required=True)
    parser.add_argument("--visualize", type=str2bool, default=False)
    args = parser.parse_args()
    return args


littletomatodonkey's avatar
littletomatodonkey 已提交
130
if __name__ == '__main__':
文幕地方's avatar
文幕地方 已提交
131 132
    args = parse_args()
    main(args)