infer_det.py 4.7 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
from ppocr.postprocess import build_post_process
37
from ppocr.utils.save_load import load_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, save_path):
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 50 51
        if not os.path.exists(save_path):
            os.makedirs(save_path)
        save_path = os.path.join(save_path, os.path.basename(img_name))
52 53 54 55
        cv2.imwrite(save_path, src_im)
        logger.info("The detected Image saved in {}".format(save_path))


Z
zhiqiu 已提交
56
@paddle.no_grad()
57
def main():
W
WenmuZhou 已提交
58 59 60 61 62
    global_config = config['Global']

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

63
    load_model(config, model)
W
WenmuZhou 已提交
64 65 66 67 68
    # build post process
    post_process_class = build_post_process(config['PostProcess'])

    # create data ops
    transforms = []
W
WenmuZhou 已提交
69
    for op in config['Eval']['dataset']['transforms']:
W
WenmuZhou 已提交
70 71 72
        op_name = list(op)[0]
        if 'Label' in op_name:
            continue
W
WenmuZhou 已提交
73
        elif op_name == 'KeepKeys':
W
WenmuZhou 已提交
74 75 76 77
            op[op_name]['keep_keys'] = ['image', 'shape']
        transforms.append(op)

    ops = create_operators(transforms, global_config)
78 79

    save_res_path = config['Global']['save_res_path']
L
LDOUBLEV 已提交
80 81 82
    if not os.path.exists(os.path.dirname(save_res_path)):
        os.makedirs(os.path.dirname(save_res_path))

W
WenmuZhou 已提交
83 84 85 86 87 88 89 90 91 92 93
    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
94
            images = paddle.to_tensor(images)
W
WenmuZhou 已提交
95 96
            preds = model(images)
            post_result = post_process_class(preds, shape_list)
L
LDOUBLEV 已提交
97 98 99

            src_img = cv2.imread(file)

W
WenmuZhou 已提交
100
            dt_boxes_json = []
L
LDOUBLEV 已提交
101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125
            # parser boxes if post_result is dict
            if isinstance(post_result, dict):
                det_box_json = {}
                for k in post_result.keys():
                    boxes = post_result[k][0]['points']
                    dt_boxes_list = []
                    for box in boxes:
                        tmp_json = {"transcription": ""}
                        tmp_json['points'] = box.tolist()
                        dt_boxes_list.append(tmp_json)
                    det_box_json[k] = dt_boxes_list
                    save_det_path = os.path.dirname(config['Global'][
                        'save_res_path']) + "/det_results_{}/".format(k)
                    draw_det_res(boxes, config, src_img, file, save_det_path)
            else:
                boxes = post_result[0]['points']
                dt_boxes_json = []
                # write result
                for box in boxes:
                    tmp_json = {"transcription": ""}
                    tmp_json['points'] = box.tolist()
                    dt_boxes_json.append(tmp_json)
                save_det_path = os.path.dirname(config['Global'][
                    'save_res_path']) + "/det_results/"
                draw_det_res(boxes, config, src_img, file, save_det_path)
W
WenmuZhou 已提交
126 127
            otstr = file + "\t" + json.dumps(dt_boxes_json) + "\n"
            fout.write(otstr.encode())
L
LDOUBLEV 已提交
128

129 130 131 132
    logger.info("success!")


if __name__ == '__main__':
W
WenmuZhou 已提交
133
    config, device, logger, vdl_writer = program.preprocess()
L
LDOUBLEV 已提交
134
    main()