predict_rec.py 13.4 KB
Newer Older
L
LDOUBLEV 已提交
1 2 3 4 5 6 7 8 9 10 11 12 13
# 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.
L
LDOUBLEV 已提交
14 15
import os
import sys
T
Topdu 已提交
16
from PIL import Image
17
__dir__ = os.path.dirname(os.path.abspath(__file__))
L
LDOUBLEV 已提交
18
sys.path.append(__dir__)
19
sys.path.append(os.path.abspath(os.path.join(__dir__, '../..')))
L
LDOUBLEV 已提交
20

L
LDOUBLEV 已提交
21 22
os.environ["FLAGS_allocator_strategy"] = 'auto_growth'

L
LDOUBLEV 已提交
23 24 25 26
import cv2
import numpy as np
import math
import time
W
WenmuZhou 已提交
27
import traceback
T
tink2123 已提交
28
import paddle
29 30

import tools.infer.utility as utility
W
WenmuZhou 已提交
31 32
from ppocr.postprocess import build_post_process
from ppocr.utils.logging import get_logger
33
from ppocr.utils.utility import get_image_file_list, check_and_read_gif
L
LDOUBLEV 已提交
34

W
WenmuZhou 已提交
35 36
logger = get_logger()

L
LDOUBLEV 已提交
37 38 39

class TextRecognizer(object):
    def __init__(self, args):
40
        self.rec_image_shape = [int(v) for v in args.rec_image_shape.split(",")]
D
dyning 已提交
41
        self.character_type = args.rec_char_type
42
        self.rec_batch_num = args.rec_batch_num
T
tink2123 已提交
43
        self.rec_algorithm = args.rec_algorithm
W
WenmuZhou 已提交
44 45
        postprocess_params = {
            'name': 'CTCLabelDecode',
T
tink2123 已提交
46
            "character_type": args.rec_char_type,
47
            "character_dict_path": args.rec_char_dict_path,
W
WenmuZhou 已提交
48
            "use_space_char": args.use_space_char
T
tink2123 已提交
49
        }
T
tink2123 已提交
50 51 52
        if self.rec_algorithm == "SRN":
            postprocess_params = {
                'name': 'SRNLabelDecode',
W
WenmuZhou 已提交
53 54 55 56 57 58 59
                "character_type": args.rec_char_type,
                "character_dict_path": args.rec_char_dict_path,
                "use_space_char": args.use_space_char
            }
        elif self.rec_algorithm == "RARE":
            postprocess_params = {
                'name': 'AttnLabelDecode',
T
tink2123 已提交
60 61 62 63
                "character_type": args.rec_char_type,
                "character_dict_path": args.rec_char_dict_path,
                "use_space_char": args.use_space_char
            }
T
Topdu 已提交
64 65 66 67 68 69 70
        elif self.rec_algorithm == 'NRTR':
            postprocess_params = {
                'name': 'NRTRLabelDecode',
                "character_type": args.rec_char_type,
                "character_dict_path": args.rec_char_dict_path,
                "use_space_char": args.use_space_char
            }
W
WenmuZhou 已提交
71
        self.postprocess_op = build_post_process(postprocess_params)
L
LDOUBLEV 已提交
72
        self.predictor, self.input_tensor, self.output_tensors, self.config = \
W
WenmuZhou 已提交
73
            utility.create_predictor(args, 'rec', logger)
T
tink2123 已提交
74 75 76 77
        self.benchmark = args.benchmark
        if args.benchmark:
            import auto_log
            pid = os.getpid()
L
LDOUBLEV 已提交
78
            gpu_id = utility.get_infer_gpuid()
T
tink2123 已提交
79 80 81
            self.autolog = auto_log.AutoLogger(
                model_name="rec",
                model_precision=args.precision,
T
tink2123 已提交
82
                batch_size=args.rec_batch_num,
T
tink2123 已提交
83
                data_shape="dynamic",
84
                save_path=None,  #args.save_log_path,
T
tink2123 已提交
85 86 87
                inference_config=self.config,
                pids=pid,
                process_name=None,
L
LDOUBLEV 已提交
88
                gpu_ids=gpu_id if args.use_gpu else None,
T
tink2123 已提交
89 90 91
                time_keys=[
                    'preprocess_time', 'inference_time', 'postprocess_time'
                ],
92 93
                warmup=2,
                logger=logger)
L
LDOUBLEV 已提交
94

95
    def resize_norm_img(self, img, max_wh_ratio):
L
LDOUBLEV 已提交
96
        imgC, imgH, imgW = self.rec_image_shape
T
Topdu 已提交
97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120
        if imgC == 1:
            img = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
            # h = img.shape[0]
            # w = img.shape[1]
            # ratio = w / float(h)
            # if math.ceil(imgH * ratio) > imgW:
            #     resized_w = imgW
            # else:
            #     resized_w = int(math.ceil(imgH * ratio))
            # resized_image = cv2.resize(img, (resized_w, imgH))
            # #norm_img = np.expand_dims(resized_image, -1)
            # #norm_img = norm_img.transpose((2, 0, 1))
            # resized_image = resized_image.astype(np.float32) / 128. - 1.
            # padding_im = np.zeros((imgC, imgH, imgW), dtype=np.float32)
            # padding_im[0, :, 0:resized_w] = resized_image

            # return padding_im
            image_pil = Image.fromarray(np.uint8(img))
            img = image_pil.resize([100, 32], Image.ANTIALIAS)
            img = np.array(img)
            norm_img = np.expand_dims(img, -1)
            norm_img = norm_img.transpose((2, 0, 1))
            return norm_img.astype(np.float32) / 128. - 1.

121
        assert imgC == img.shape[2]
T
tink2123 已提交
122 123
        max_wh_ratio = max(max_wh_ratio, imgW / imgH)
        imgW = int((32 * max_wh_ratio))
124
        h, w = img.shape[:2]
125 126 127 128 129
        ratio = w / float(h)
        if math.ceil(imgH * ratio) > imgW:
            resized_w = imgW
        else:
            resized_w = int(math.ceil(imgH * ratio))
T
tink2123 已提交
130
        resized_image = cv2.resize(img, (resized_w, imgH))
L
LDOUBLEV 已提交
131 132 133 134 135 136 137 138
        resized_image = resized_image.astype('float32')
        resized_image = resized_image.transpose((2, 0, 1)) / 255
        resized_image -= 0.5
        resized_image /= 0.5
        padding_im = np.zeros((imgC, imgH, imgW), dtype=np.float32)
        padding_im[:, :, 0:resized_w] = resized_image
        return padding_im

T
tink2123 已提交
139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210
    def resize_norm_img_srn(self, img, image_shape):
        imgC, imgH, imgW = image_shape

        img_black = np.zeros((imgH, imgW))
        im_hei = img.shape[0]
        im_wid = img.shape[1]

        if im_wid <= im_hei * 1:
            img_new = cv2.resize(img, (imgH * 1, imgH))
        elif im_wid <= im_hei * 2:
            img_new = cv2.resize(img, (imgH * 2, imgH))
        elif im_wid <= im_hei * 3:
            img_new = cv2.resize(img, (imgH * 3, imgH))
        else:
            img_new = cv2.resize(img, (imgW, imgH))

        img_np = np.asarray(img_new)
        img_np = cv2.cvtColor(img_np, cv2.COLOR_BGR2GRAY)
        img_black[:, 0:img_np.shape[1]] = img_np
        img_black = img_black[:, :, np.newaxis]

        row, col, c = img_black.shape
        c = 1

        return np.reshape(img_black, (c, row, col)).astype(np.float32)

    def srn_other_inputs(self, image_shape, num_heads, max_text_length):

        imgC, imgH, imgW = image_shape
        feature_dim = int((imgH / 8) * (imgW / 8))

        encoder_word_pos = np.array(range(0, feature_dim)).reshape(
            (feature_dim, 1)).astype('int64')
        gsrm_word_pos = np.array(range(0, max_text_length)).reshape(
            (max_text_length, 1)).astype('int64')

        gsrm_attn_bias_data = np.ones((1, max_text_length, max_text_length))
        gsrm_slf_attn_bias1 = np.triu(gsrm_attn_bias_data, 1).reshape(
            [-1, 1, max_text_length, max_text_length])
        gsrm_slf_attn_bias1 = np.tile(
            gsrm_slf_attn_bias1,
            [1, num_heads, 1, 1]).astype('float32') * [-1e9]

        gsrm_slf_attn_bias2 = np.tril(gsrm_attn_bias_data, -1).reshape(
            [-1, 1, max_text_length, max_text_length])
        gsrm_slf_attn_bias2 = np.tile(
            gsrm_slf_attn_bias2,
            [1, num_heads, 1, 1]).astype('float32') * [-1e9]

        encoder_word_pos = encoder_word_pos[np.newaxis, :]
        gsrm_word_pos = gsrm_word_pos[np.newaxis, :]

        return [
            encoder_word_pos, gsrm_word_pos, gsrm_slf_attn_bias1,
            gsrm_slf_attn_bias2
        ]

    def process_image_srn(self, img, image_shape, num_heads, max_text_length):
        norm_img = self.resize_norm_img_srn(img, image_shape)
        norm_img = norm_img[np.newaxis, :]

        [encoder_word_pos, gsrm_word_pos, gsrm_slf_attn_bias1, gsrm_slf_attn_bias2] = \
            self.srn_other_inputs(image_shape, num_heads, max_text_length)

        gsrm_slf_attn_bias1 = gsrm_slf_attn_bias1.astype(np.float32)
        gsrm_slf_attn_bias2 = gsrm_slf_attn_bias2.astype(np.float32)
        encoder_word_pos = encoder_word_pos.astype(np.int64)
        gsrm_word_pos = gsrm_word_pos.astype(np.int64)

        return (norm_img, encoder_word_pos, gsrm_word_pos, gsrm_slf_attn_bias1,
                gsrm_slf_attn_bias2)

L
LDOUBLEV 已提交
211 212
    def __call__(self, img_list):
        img_num = len(img_list)
213
        # Calculate the aspect ratio of all text bars
214 215 216
        width_list = []
        for img in img_list:
            width_list.append(img.shape[1] / float(img.shape[0]))
张欣-男's avatar
张欣-男 已提交
217
        # Sorting can speed up the recognition process
218 219
        indices = np.argsort(np.array(width_list))
        rec_res = [['', 0.0]] * img_num
220
        batch_num = self.rec_batch_num
L
LDOUBLEV 已提交
221
        st = time.time()
T
tink2123 已提交
222 223
        if self.benchmark:
            self.autolog.times.start()
L
LDOUBLEV 已提交
224 225 226
        for beg_img_no in range(0, img_num, batch_num):
            end_img_no = min(img_num, beg_img_no + batch_num)
            norm_img_batch = []
227
            max_wh_ratio = 0
L
LDOUBLEV 已提交
228
            for ino in range(beg_img_no, end_img_no):
229
                h, w = img_list[indices[ino]].shape[0:2]
230 231 232
                wh_ratio = w * 1.0 / h
                max_wh_ratio = max(max_wh_ratio, wh_ratio)
            for ino in range(beg_img_no, end_img_no):
T
tink2123 已提交
233 234 235 236 237 238
                if self.rec_algorithm != "SRN":
                    norm_img = self.resize_norm_img(img_list[indices[ino]],
                                                    max_wh_ratio)
                    norm_img = norm_img[np.newaxis, :]
                    norm_img_batch.append(norm_img)
                else:
L
LDOUBLEV 已提交
239 240
                    norm_img = self.process_image_srn(
                        img_list[indices[ino]], self.rec_image_shape, 8, 25)
T
tink2123 已提交
241 242 243 244 245 246 247 248 249
                    encoder_word_pos_list = []
                    gsrm_word_pos_list = []
                    gsrm_slf_attn_bias1_list = []
                    gsrm_slf_attn_bias2_list = []
                    encoder_word_pos_list.append(norm_img[1])
                    gsrm_word_pos_list.append(norm_img[2])
                    gsrm_slf_attn_bias1_list.append(norm_img[3])
                    gsrm_slf_attn_bias2_list.append(norm_img[4])
                    norm_img_batch.append(norm_img[0])
L
LDOUBLEV 已提交
250 251
            norm_img_batch = np.concatenate(norm_img_batch)
            norm_img_batch = norm_img_batch.copy()
T
tink2123 已提交
252 253
            if self.benchmark:
                self.autolog.times.stamp()
T
tink2123 已提交
254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279

            if self.rec_algorithm == "SRN":
                encoder_word_pos_list = np.concatenate(encoder_word_pos_list)
                gsrm_word_pos_list = np.concatenate(gsrm_word_pos_list)
                gsrm_slf_attn_bias1_list = np.concatenate(
                    gsrm_slf_attn_bias1_list)
                gsrm_slf_attn_bias2_list = np.concatenate(
                    gsrm_slf_attn_bias2_list)

                inputs = [
                    norm_img_batch,
                    encoder_word_pos_list,
                    gsrm_word_pos_list,
                    gsrm_slf_attn_bias1_list,
                    gsrm_slf_attn_bias2_list,
                ]
                input_names = self.predictor.get_input_names()
                for i in range(len(input_names)):
                    input_tensor = self.predictor.get_input_handle(input_names[
                        i])
                    input_tensor.copy_from_cpu(inputs[i])
                self.predictor.run()
                outputs = []
                for output_tensor in self.output_tensors:
                    output = output_tensor.copy_to_cpu()
                    outputs.append(output)
T
tink2123 已提交
280 281
                if self.benchmark:
                    self.autolog.times.stamp()
T
tink2123 已提交
282 283 284 285 286 287 288 289
                preds = {"predict": outputs[2]}
            else:
                self.input_tensor.copy_from_cpu(norm_img_batch)
                self.predictor.run()
                outputs = []
                for output_tensor in self.output_tensors:
                    output = output_tensor.copy_to_cpu()
                    outputs.append(output)
T
tink2123 已提交
290 291
                if self.benchmark:
                    self.autolog.times.stamp()
T
Topdu 已提交
292 293 294 295
                if len(outputs) != 1:
                    preds = outputs
                else:
                    preds = outputs[0]
W
WenmuZhou 已提交
296 297 298
            rec_result = self.postprocess_op(preds)
            for rno in range(len(rec_result)):
                rec_res[indices[beg_img_no + rno]] = rec_result[rno]
T
tink2123 已提交
299 300
            if self.benchmark:
                self.autolog.times.end(stamp=True)
L
LDOUBLEV 已提交
301
        return rec_res, time.time() - st
L
LDOUBLEV 已提交
302 303


304
def main(args):
D
dyning 已提交
305
    image_file_list = get_image_file_list(args.image_dir)
L
LDOUBLEV 已提交
306 307 308
    text_recognizer = TextRecognizer(args)
    valid_image_file_list = []
    img_list = []
L
LDOUBLEV 已提交
309

310
    # warmup 2 times
L
LDOUBLEV 已提交
311 312
    if args.warmup:
        img = np.random.uniform(0, 255, [32, 320, 3]).astype(np.uint8)
313
        for i in range(2):
L
LDOUBLEV 已提交
314
            res = text_recognizer([img] * int(args.rec_batch_num))
L
LDOUBLEV 已提交
315

L
LDOUBLEV 已提交
316
    for image_file in image_file_list:
L
LDOUBLEV 已提交
317 318 319
        img, flag = check_and_read_gif(image_file)
        if not flag:
            img = cv2.imread(image_file)
L
LDOUBLEV 已提交
320 321 322 323 324
        if img is None:
            logger.info("error in loading image:{}".format(image_file))
            continue
        valid_image_file_list.append(image_file)
        img_list.append(img)
L
LDOUBLEV 已提交
325 326 327 328 329 330 331 332 333 334
    try:
        rec_res, _ = text_recognizer(img_list)

    except Exception as E:
        logger.info(traceback.format_exc())
        logger.info(E)
        exit()
    for ino in range(len(img_list)):
        logger.info("Predicts of {}:{}".format(valid_image_file_list[ino],
                                               rec_res[ino]))
T
tink2123 已提交
335 336
    if args.benchmark:
        text_recognizer.autolog.report()
337 338 339 340


if __name__ == "__main__":
    main(utility.parse_args())