infer_det.py 3.8 KB
Newer Older
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
# 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.

from __future__ import absolute_import
from __future__ import division
from __future__ import print_function

import numpy as np

L
LDOUBLEV 已提交
21 22
import os
import sys
W
WenmuZhou 已提交
23

24
__dir__ = os.path.dirname(os.path.abspath(__file__))
L
LDOUBLEV 已提交
25
sys.path.append(__dir__)
26
sys.path.append(os.path.abspath(os.path.join(__dir__, '..')))
27

L
LDOUBLEV 已提交
28 29
os.environ["FLAGS_allocator_strategy"] = 'auto_growth'

L
LDOUBLEV 已提交
30
import cv2
W
WenmuZhou 已提交
31 32
import json
import paddle
33

W
WenmuZhou 已提交
34
from ppocr.data import create_operators, transform
W
WenmuZhou 已提交
35
from ppocr.modeling.architectures import build_model
W
WenmuZhou 已提交
36 37
from ppocr.postprocess import build_post_process
from ppocr.utils.save_load import init_model
W
WenmuZhou 已提交
38
from ppocr.utils.utility import get_image_file_list
W
WenmuZhou 已提交
39
import tools.program as program
40 41


L
LDOUBLEV 已提交
42
def draw_det_res(dt_boxes, config, img, img_name):
43 44
    if len(dt_boxes) > 0:
        import cv2
L
LDOUBLEV 已提交
45
        src_im = img
46 47 48
        for box in dt_boxes:
            box = box.astype(np.int32).reshape((-1, 1, 2))
            cv2.polylines(src_im, [box], True, color=(255, 255, 0), thickness=2)
L
LDOUBLEV 已提交
49
        save_det_path = os.path.dirname(config['Global'][
50 51 52
            'save_res_path']) + "/det_results/"
        if not os.path.exists(save_det_path):
            os.makedirs(save_det_path)
L
LDOUBLEV 已提交
53
        save_path = os.path.join(save_det_path, os.path.basename(img_name))
54 55 56 57 58
        cv2.imwrite(save_path, src_im)
        logger.info("The detected Image saved in {}".format(save_path))


def main():
W
WenmuZhou 已提交
59 60 61 62 63 64 65 66 67 68 69 70
    global_config = config['Global']

    # build model
    model = build_model(config['Architecture'])

    init_model(config, model, logger)

    # build post process
    post_process_class = build_post_process(config['PostProcess'])

    # create data ops
    transforms = []
W
WenmuZhou 已提交
71
    for op in config['Eval']['dataset']['transforms']:
W
WenmuZhou 已提交
72 73 74
        op_name = list(op)[0]
        if 'Label' in op_name:
            continue
W
WenmuZhou 已提交
75
        elif op_name == 'KeepKeys':
W
WenmuZhou 已提交
76 77 78 79
            op[op_name]['keep_keys'] = ['image', 'shape']
        transforms.append(op)

    ops = create_operators(transforms, global_config)
80 81

    save_res_path = config['Global']['save_res_path']
L
LDOUBLEV 已提交
82 83 84
    if not os.path.exists(os.path.dirname(save_res_path)):
        os.makedirs(os.path.dirname(save_res_path))

W
WenmuZhou 已提交
85 86 87 88 89 90 91 92 93 94 95
    model.eval()
    with open(save_res_path, "wb") as fout:
        for file in get_image_file_list(config['Global']['infer_img']):
            logger.info("infer_img: {}".format(file))
            with open(file, 'rb') as f:
                img = f.read()
                data = {'image': img}
            batch = transform(data, ops)

            images = np.expand_dims(batch[0], axis=0)
            shape_list = np.expand_dims(batch[1], axis=0)
W
WenmuZhou 已提交
96
            images = paddle.to_tensor(images)
W
WenmuZhou 已提交
97 98 99 100 101 102 103 104 105 106 107 108 109
            preds = model(images)
            post_result = post_process_class(preds, shape_list)
            boxes = post_result[0]['points']
            # write resule
            dt_boxes_json = []
            for box in boxes:
                tmp_json = {"transcription": ""}
                tmp_json['points'] = box.tolist()
                dt_boxes_json.append(tmp_json)
            otstr = file + "\t" + json.dumps(dt_boxes_json) + "\n"
            fout.write(otstr.encode())
            src_img = cv2.imread(file)
            draw_det_res(boxes, config, src_img, file)
110 111 112 113
    logger.info("success!")


if __name__ == '__main__':
W
WenmuZhou 已提交
114 115
    config, device, logger, vdl_writer = program.preprocess()
    main()