Skip to content

πŸ“„ Awesome OCR multiple programing languages toolkits based on ONNX Runtime, OpenVINO, MNN, PaddlePaddle, TensorRT and PyTorch.

整理代码

SWHLd8c45c8a0c69
diff --git a/python/ch_ppocr_mobile_v2_cls/text_cls.py b/python/ch_ppocr_mobile_v2_cls/text_cls.py
index b226403..8992ec4 100755
--- a/python/ch_ppocr_mobile_v2_cls/text_cls.py
+++ b/python/ch_ppocr_mobile_v2_cls/text_cls.py
@@ -121,12 +121,12 @@ class TextClassifier(object):
if __name__ == "__main__":
parser = argparse.ArgumentParser()
- parser.add_argument('--image_dir', type=str, help='image_dir|image_path')
- parser.add_argument('--cls_model_dir', type=str, help='cls_model_dir')
+ parser.add_argument('--image_path', type=str, help='image_dir|image_path')
+ parser.add_argument('--model_path', type=str, help='model_path')
args = parser.parse_args()
- text_classifier = TextClassifier(args.cls_model_dir)
+ text_classifier = TextClassifier(args.model_path)
- img_list, cls_res, predict_time = text_classifier(args.image_dir)
+ img_list, cls_res, predict_time = text_classifier(args.image_path)
for ino in range(len(img_list)):
print(f"εˆ†η±»η»“ζžœ:{cls_res[ino]}")
diff --git a/python/ch_ppocr_mobile_v2_det/text_detect.py b/python/ch_ppocr_mobile_v2_det/text_detect.py
index 0bf6163..e02fbff 100755
--- a/python/ch_ppocr_mobile_v2_det/text_detect.py
+++ b/python/ch_ppocr_mobile_v2_det/text_detect.py
@@ -137,10 +137,8 @@ class TextDetector(object):
if __name__ == "__main__":
parser = argparse.ArgumentParser()
- parser.add_argument('--image_path', type=str, default=None,
- help='image_path|image_dir')
- parser.add_argument('--model_path', type=str, default=None,
- help='model_path')
+ parser.add_argument('--image_path', type=str, help='image_path|image_dir')
+ parser.add_argument('--model_path', type=str, help='model_path')
args = parser.parse_args()
text_detector = TextDetector(args.model_path)
diff --git a/python/ch_ppocr_mobile_v2_rec/text_recognize.py b/python/ch_ppocr_mobile_v2_rec/text_recognize.py
index 42a63db..53ec726 100755
--- a/python/ch_ppocr_mobile_v2_rec/text_recognize.py
+++ b/python/ch_ppocr_mobile_v2_rec/text_recognize.py
@@ -32,23 +32,19 @@ root_path = Path(__file__).parent
class TextRecognizer(object):
def __init__(self, rec_model_path):
self.rec_image_shape = [3, 32, 320]
- self.character_type = 'ch'
self.rec_batch_num = 6
- self.rec_algorithm = 'CRNN'
self.character_dict_path = str(root_path / 'ppocr_keys_v1.txt')
- self.use_space_char = True
- self.postprocess_op = CTCLabelDecode(self.character_dict_path,
- self.character_type,
- self.use_space_char)
+ self.postprocess_op = CTCLabelDecode(self.character_dict_path)
self.session = onnxruntime.InferenceSession(rec_model_path)
def resize_norm_img(self, img, max_wh_ratio):
img_channel, img_height, img_width = self.rec_image_shape
assert img_channel == img.shape[2]
- if self.character_type == "ch":
- img_width = int((32 * max_wh_ratio))
- max_wh_ratio = 1
+
+ img_width = int((32 * max_wh_ratio))
+ max_wh_ratio = 1
+
h, w = img.shape[:2]
ratio = w / float(h)
if math.ceil(img_height * ratio) > img_width:
@@ -127,10 +123,10 @@ class TextRecognizer(object):
if __name__ == "__main__":
parser = argparse.ArgumentParser()
- parser.add_argument('image_dir', type=str, help='image_dir|image_path')
- parser.add_argument('model_path', type=str, help='rec_model_path')
+ parser.add_argument('--image_path', type=str, help='image_dir|image_path')
+ parser.add_argument('--model_path', type=str, help='rec_model_path')
args = parser.parse_args()
text_recognizer = TextRecognizer(args.model_path)
- rec_res, predict_time = text_recognizer(args.image_dir)
+ rec_res, predict_time = text_recognizer(args.image_path)
print(f'θ―†εˆ«η»“ζžœ: {rec_res}\t cost: {predict_time}s')
diff --git a/python/ch_ppocr_mobile_v2_rec/utils.py b/python/ch_ppocr_mobile_v2_rec/utils.py
index 8836d71..d3b2c65 100755
--- a/python/ch_ppocr_mobile_v2_rec/utils.py
+++ b/python/ch_ppocr_mobile_v2_rec/utils.py
@@ -1,7 +1,7 @@
# !/usr/bin/env python
# -*- encoding: utf-8 -*-
-# @File: PostProcess.py
-# @Time: 2021/01/19 14:12:35
+# @File: utils.py
+# @Time: 2021/05/22 10:53:17
# @Author: Max
import cv2
import os
@@ -9,35 +9,75 @@ import imghdr
import numpy as np
-class BaseRecLabelDecode(object):
+def get_image_file_list(img_file):
+ imgs_lists = []
+ if img_file is None or not os.path.exists(img_file):
+ raise FileNotFoundError("not found any img file in {}".format(img_file))
+
+ img_end = {'jpg', 'bmp', 'png', 'jpeg', 'rgb', 'tif', 'tiff', 'gif', 'GIF'}
+ if os.path.isfile(img_file) and imghdr.what(img_file) in img_end:
+ imgs_lists.append(img_file)
+ elif os.path.isdir(img_file):
+ for single_file in os.listdir(img_file):
+ file_path = os.path.join(img_file, single_file)
+ if imghdr.what(file_path) in img_end:
+ imgs_lists.append(file_path)
+ if len(imgs_lists) == 0:
+ raise FileNotFoundError("not found any img file in {}".format(img_file))
+ return imgs_lists
+
+
+def check_and_read_gif(img_path):
+ if os.path.basename(img_path)[-3:] in ['gif', 'GIF']:
+ gif = cv2.VideoCapture(img_path)
+ ret, frame = gif.read()
+ if not ret:
+ print("Cannot read {}. This gif image maybe corrupted.")
+ return None, False
+ if len(frame.shape) == 2 or frame.shape[-1] == 1:
+ frame = cv2.cvtColor(frame, cv2.COLOR_GRAY2RGB)
+ imgvalue = frame[:, :, ::-1]
+ return imgvalue, True
+ return None, False
+
+
+class CTCLabelDecode(object):
""" Convert between text-label and text-index """
- def __init__(self,
- character_dict_path=None,
- character_type='ch',
- use_space_char=False):
- self.character_str = ""
- assert character_dict_path is not None, "character_dict_path should not be None when character_type is ch"
+ def __init__(self, character_dict_path):
+ super(CTCLabelDecode, self).__init__()
+ self.character_str = []
+ assert character_dict_path is not None, "character_dict_path should not be None"
with open(character_dict_path, "rb") as fin:
lines = fin.readlines()
for line in lines:
line = line.decode('utf-8').strip("\n").strip("\r\n")
- self.character_str += line
- if use_space_char:
- self.character_str += " "
- dict_character = list(self.character_str)
+ self.character_str.append(line)
+ dict_character = self.character_str
- self.character_type = character_type
dict_character = self.add_special_char(dict_character)
self.dict = {}
for i, char in enumerate(dict_character):
self.dict[char] = i
self.character = dict_character
+ def __call__(self, preds, label=None, *args, **kwargs):
+ preds_idx = preds.argmax(axis=2)
+ preds_prob = preds.max(axis=2)
+ text = self.decode(preds_idx, preds_prob, is_remove_duplicate=True)
+ if label is None:
+ return text
+ label = self.decode(label)
+ return text, label
+
def add_special_char(self, dict_character):
+ dict_character = ['blank'] + dict_character
return dict_character
- def decode(self, text_index, text_prob=None, is_remove_duplicate=True):
+ def get_ignored_tokens(self):
+ return [0] # for ctc blank
+
+ def decode(self, text_index, text_prob=None, is_remove_duplicate=False):
""" convert text-index into text-label. """
result_list = []
ignored_tokens = self.get_ignored_tokens()
@@ -63,62 +103,3 @@ class BaseRecLabelDecode(object):
result_list.append((text, np.mean(conf_list)))
return result_list
- def get_ignored_tokens(self):
- return [0] # for ctc blank
-
-
-class CTCLabelDecode(BaseRecLabelDecode):
- """ Convert between text-label and text-index """
-
- def __init__(self,
- character_dict_path=None,
- character_type='ch',
- use_space_char=False,
- **kwargs):
- super(CTCLabelDecode, self).__init__(character_dict_path,
- character_type, use_space_char)
-
- def __call__(self, preds, label=None, *args, **kwargs):
- preds_idx = preds.argmax(axis=2)
- preds_prob = preds.max(axis=2)
- text = self.decode(preds_idx, preds_prob)
- if label is None:
- return text
- label = self.decode(label, is_remove_duplicate=False)
- return text, label
-
- def add_special_char(self, dict_character):
- dict_character = ['blank'] + dict_character
- return dict_character
-
-
-def get_image_file_list(img_file):
- imgs_lists = []
- if img_file is None or not os.path.exists(img_file):
- raise Exception("not found any img file in {}".format(img_file))
-
- img_end = {'jpg', 'bmp', 'png', 'jpeg', 'rgb', 'tif', 'tiff', 'gif', 'GIF'}
- if os.path.isfile(img_file) and imghdr.what(img_file) in img_end:
- imgs_lists.append(img_file)
- elif os.path.isdir(img_file):
- for single_file in os.listdir(img_file):
- file_path = os.path.join(img_file, single_file)
- if imghdr.what(file_path) in img_end:
- imgs_lists.append(file_path)
- if len(imgs_lists) == 0:
- raise Exception("not found any img file in {}".format(img_file))
- return imgs_lists
-
-
-def check_and_read_gif(img_path):
- if os.path.basename(img_path)[-3:] in ['gif', 'GIF']:
- gif = cv2.VideoCapture(img_path)
- ret, frame = gif.read()
- if not ret:
- print("Cannot read {}. This gif image maybe corrupted.")
- return None, False
- if len(frame.shape) == 2 or frame.shape[-1] == 1:
- frame = cv2.cvtColor(frame, cv2.COLOR_GRAY2RGB)
- imgvalue = frame[:, :, ::-1]
- return imgvalue, True
- return None, False
\ No newline at end of file
diff --git a/python/ch_ppocr_server_v2_det/text_detect.py b/python/ch_ppocr_server_v2_det/text_detect.py
index a37ad41..a342a8c 100755
--- a/python/ch_ppocr_server_v2_det/text_detect.py
+++ b/python/ch_ppocr_server_v2_det/text_detect.py
@@ -135,10 +135,8 @@ class TextDetector(object):
if __name__ == "__main__":
parser = argparse.ArgumentParser()
- parser.add_argument('--image_path', type=str, default=None,
- help='image_path|image_dir')
- parser.add_argument('--model_path', type=str, default=None,
- help='model_path')
+ parser.add_argument('--image_path', type=str, help='image_path|image_dir')
+ parser.add_argument('--model_path', type=str, help='model_path')
args = parser.parse_args()
text_detector = TextDetector(args.model_path)
diff --git a/python/ch_ppocr_server_v2_rec/text_recognize.py b/python/ch_ppocr_server_v2_rec/text_recognize.py
index af7dbd8..5f1e799 100755
--- a/python/ch_ppocr_server_v2_rec/text_recognize.py
+++ b/python/ch_ppocr_server_v2_rec/text_recognize.py
@@ -32,23 +32,19 @@ root_path = Path(__file__).parent
class TextRecognizer(object):
def __init__(self, rec_model_path):
self.rec_image_shape = [3, 32, 320]
- self.character_type = 'ch'
self.rec_batch_num = 6
- self.rec_algorithm = 'CRNN'
self.character_dict_path = str(root_path / 'ppocr_keys_v1.txt')
- self.use_space_char = True
- self.postprocess_op = CTCLabelDecode(self.character_dict_path,
- self.character_type,
- self.use_space_char)
+ self.postprocess_op = CTCLabelDecode(self.character_dict_path)
self.session = onnxruntime.InferenceSession(rec_model_path)
def resize_norm_img(self, img, max_wh_ratio):
img_channel, img_height, img_width = self.rec_image_shape
assert img_channel == img.shape[2]
- if self.character_type == "ch":
- img_width = int((32 * max_wh_ratio))
- max_wh_ratio = 1
+
+ img_width = int((32 * max_wh_ratio))
+ max_wh_ratio = 1
+
h, w = img.shape[:2]
ratio = w / float(h)
if math.ceil(img_height * ratio) > img_width:
@@ -128,10 +124,10 @@ class TextRecognizer(object):
if __name__ == "__main__":
parser = argparse.ArgumentParser()
- parser.add_argument('--image_dir', type=str, help='image_dir|image_path')
+ parser.add_argument('--image_path', type=str, help='image_dir|image_path')
parser.add_argument('--model_path', type=str, help='rec_model_path')
args = parser.parse_args()
text_recognizer = TextRecognizer(args.model_path)
- rec_res, predict_time = text_recognizer(args.image_dir)
+ rec_res, predict_time = text_recognizer(args.image_path)
print(f'θ―†εˆ«η»“ζžœ: {rec_res}\t cost: {predict_time}s')
diff --git a/python/ch_ppocr_server_v2_rec/utils.py b/python/ch_ppocr_server_v2_rec/utils.py
index 8836d71..0831c37 100755
--- a/python/ch_ppocr_server_v2_rec/utils.py
+++ b/python/ch_ppocr_server_v2_rec/utils.py
@@ -9,35 +9,75 @@ import imghdr
import numpy as np
-class BaseRecLabelDecode(object):
+def get_image_file_list(img_file):
+ imgs_lists = []
+ if img_file is None or not os.path.exists(img_file):
+ raise FileNotFoundError("not found any img file in {}".format(img_file))
+
+ img_end = {'jpg', 'bmp', 'png', 'jpeg', 'rgb', 'tif', 'tiff', 'gif', 'GIF'}
+ if os.path.isfile(img_file) and imghdr.what(img_file) in img_end:
+ imgs_lists.append(img_file)
+ elif os.path.isdir(img_file):
+ for single_file in os.listdir(img_file):
+ file_path = os.path.join(img_file, single_file)
+ if imghdr.what(file_path) in img_end:
+ imgs_lists.append(file_path)
+ if len(imgs_lists) == 0:
+ raise FileNotFoundError("not found any img file in {}".format(img_file))
+ return imgs_lists
+
+
+def check_and_read_gif(img_path):
+ if os.path.basename(img_path)[-3:] in ['gif', 'GIF']:
+ gif = cv2.VideoCapture(img_path)
+ ret, frame = gif.read()
+ if not ret:
+ print("Cannot read {}. This gif image maybe corrupted.")
+ return None, False
+ if len(frame.shape) == 2 or frame.shape[-1] == 1:
+ frame = cv2.cvtColor(frame, cv2.COLOR_GRAY2RGB)
+ imgvalue = frame[:, :, ::-1]
+ return imgvalue, True
+ return None, False
+
+
+class CTCLabelDecode(object):
""" Convert between text-label and text-index """
- def __init__(self,
- character_dict_path=None,
- character_type='ch',
- use_space_char=False):
- self.character_str = ""
- assert character_dict_path is not None, "character_dict_path should not be None when character_type is ch"
+ def __init__(self, character_dict_path):
+ super(CTCLabelDecode, self).__init__()
+ self.character_str = []
+ assert character_dict_path is not None, "character_dict_path should not be None"
with open(character_dict_path, "rb") as fin:
lines = fin.readlines()
for line in lines:
line = line.decode('utf-8').strip("\n").strip("\r\n")
- self.character_str += line
- if use_space_char:
- self.character_str += " "
- dict_character = list(self.character_str)
+ self.character_str.append(line)
+ dict_character = self.character_str
- self.character_type = character_type
dict_character = self.add_special_char(dict_character)
self.dict = {}
for i, char in enumerate(dict_character):
self.dict[char] = i
self.character = dict_character
+ def __call__(self, preds, label=None, *args, **kwargs):
+ preds_idx = preds.argmax(axis=2)
+ preds_prob = preds.max(axis=2)
+ text = self.decode(preds_idx, preds_prob, is_remove_duplicate=True)
+ if label is None:
+ return text
+ label = self.decode(label)
+ return text, label
+
def add_special_char(self, dict_character):
+ dict_character = ['blank'] + dict_character
return dict_character
- def decode(self, text_index, text_prob=None, is_remove_duplicate=True):
+ def get_ignored_tokens(self):
+ return [0] # for ctc blank
+
+ def decode(self, text_index, text_prob=None, is_remove_duplicate=False):
""" convert text-index into text-label. """
result_list = []
ignored_tokens = self.get_ignored_tokens()
@@ -63,62 +103,3 @@ class BaseRecLabelDecode(object):
result_list.append((text, np.mean(conf_list)))
return result_list
- def get_ignored_tokens(self):
- return [0] # for ctc blank
-
-
-class CTCLabelDecode(BaseRecLabelDecode):
- """ Convert between text-label and text-index """
-
- def __init__(self,
- character_dict_path=None,
- character_type='ch',
- use_space_char=False,
- **kwargs):
- super(CTCLabelDecode, self).__init__(character_dict_path,
- character_type, use_space_char)
-
- def __call__(self, preds, label=None, *args, **kwargs):
- preds_idx = preds.argmax(axis=2)
- preds_prob = preds.max(axis=2)
- text = self.decode(preds_idx, preds_prob)
- if label is None:
- return text
- label = self.decode(label, is_remove_duplicate=False)
- return text, label
-
- def add_special_char(self, dict_character):
- dict_character = ['blank'] + dict_character
- return dict_character
-
-
-def get_image_file_list(img_file):
- imgs_lists = []
- if img_file is None or not os.path.exists(img_file):
- raise Exception("not found any img file in {}".format(img_file))
-
- img_end = {'jpg', 'bmp', 'png', 'jpeg', 'rgb', 'tif', 'tiff', 'gif', 'GIF'}
- if os.path.isfile(img_file) and imghdr.what(img_file) in img_end:
- imgs_lists.append(img_file)
- elif os.path.isdir(img_file):
- for single_file in os.listdir(img_file):
- file_path = os.path.join(img_file, single_file)
- if imghdr.what(file_path) in img_end:
- imgs_lists.append(file_path)
- if len(imgs_lists) == 0:
- raise Exception("not found any img file in {}".format(img_file))
- return imgs_lists
-
-
-def check_and_read_gif(img_path):
- if os.path.basename(img_path)[-3:] in ['gif', 'GIF']:
- gif = cv2.VideoCapture(img_path)
- ret, frame = gif.read()
- if not ret:
- print("Cannot read {}. This gif image maybe corrupted.")
- return None, False
- if len(frame.shape) == 2 or frame.shape[-1] == 1:
- frame = cv2.cvtColor(frame, cv2.COLOR_GRAY2RGB)
- imgvalue = frame[:, :, ::-1]
- return imgvalue, True
- return None, False
\ No newline at end of file
diff --git a/python/japan_ppocr_mobile_v2_rec/text_recognize.py b/python/japan_ppocr_mobile_v2_rec/text_recognize.py
index 9836383..ec398b3 100644
--- a/python/japan_ppocr_mobile_v2_rec/text_recognize.py
+++ b/python/japan_ppocr_mobile_v2_rec/text_recognize.py
@@ -32,23 +32,19 @@ root_path = Path(__file__).parent
class TextRecognizer(object):
def __init__(self, rec_model_path):
self.rec_image_shape = [3, 32, 320]
- self.character_type = 'ch'
self.rec_batch_num = 6
- self.rec_algorithm = 'CRNN'
self.character_dict_path = str(root_path / 'japan_dict.txt')
- self.use_space_char = True
- self.postprocess_op = CTCLabelDecode(self.character_dict_path,
- self.character_type,
- self.use_space_char)
+ self.postprocess_op = CTCLabelDecode(self.character_dict_path)
self.session = onnxruntime.InferenceSession(rec_model_path)
def resize_norm_img(self, img, max_wh_ratio):
img_channel, img_height, img_width = self.rec_image_shape
assert img_channel == img.shape[2]
- if self.character_type == "ch":
- img_width = int((32 * max_wh_ratio))
- max_wh_ratio = 1
+
+ img_width = int((32 * max_wh_ratio))
+ max_wh_ratio = 1
+
h, w = img.shape[:2]
ratio = w / float(h)
if math.ceil(img_height * ratio) > img_width:
@@ -127,10 +123,10 @@ class TextRecognizer(object):
if __name__ == "__main__":
parser = argparse.ArgumentParser()
- parser.add_argument('--image_dir', type=str, help='image_dir|image_path')
+ parser.add_argument('--image_path', type=str, help='image_dir|image_path')
parser.add_argument('--model_path', type=str, help='rec_model_path')
args = parser.parse_args()
text_recognizer = TextRecognizer(args.model_path)
- rec_res, predict_time = text_recognizer(args.image_dir)
+ rec_res, predict_time = text_recognizer(args.image_path)
print(f'θ―†εˆ«η»“ζžœ: {rec_res}\t cost: {predict_time}s')
diff --git a/python/japan_ppocr_mobile_v2_rec/utils.py b/python/japan_ppocr_mobile_v2_rec/utils.py
index 8836d71..d3b2c65 100644
--- a/python/japan_ppocr_mobile_v2_rec/utils.py
+++ b/python/japan_ppocr_mobile_v2_rec/utils.py
@@ -1,7 +1,7 @@
# !/usr/bin/env python
# -*- encoding: utf-8 -*-
-# @File: PostProcess.py
-# @Time: 2021/01/19 14:12:35
+# @File: utils.py
+# @Time: 2021/05/22 10:53:17
# @Author: Max
import cv2
import os
@@ -9,35 +9,75 @@ import imghdr
import numpy as np
-class BaseRecLabelDecode(object):
+def get_image_file_list(img_file):
+ imgs_lists = []
+ if img_file is None or not os.path.exists(img_file):
+ raise FileNotFoundError("not found any img file in {}".format(img_file))
+
+ img_end = {'jpg', 'bmp', 'png', 'jpeg', 'rgb', 'tif', 'tiff', 'gif', 'GIF'}
+ if os.path.isfile(img_file) and imghdr.what(img_file) in img_end:
+ imgs_lists.append(img_file)
+ elif os.path.isdir(img_file):
+ for single_file in os.listdir(img_file):
+ file_path = os.path.join(img_file, single_file)
+ if imghdr.what(file_path) in img_end:
+ imgs_lists.append(file_path)
+ if len(imgs_lists) == 0:
+ raise FileNotFoundError("not found any img file in {}".format(img_file))
+ return imgs_lists
+
+
+def check_and_read_gif(img_path):
+ if os.path.basename(img_path)[-3:] in ['gif', 'GIF']:
+ gif = cv2.VideoCapture(img_path)
+ ret, frame = gif.read()
+ if not ret:
+ print("Cannot read {}. This gif image maybe corrupted.")
+ return None, False
+ if len(frame.shape) == 2 or frame.shape[-1] == 1:
+ frame = cv2.cvtColor(frame, cv2.COLOR_GRAY2RGB)
+ imgvalue = frame[:, :, ::-1]
+ return imgvalue, True
+ return None, False
+
+
+class CTCLabelDecode(object):
""" Convert between text-label and text-index """
- def __init__(self,
- character_dict_path=None,
- character_type='ch',
- use_space_char=False):
- self.character_str = ""
- assert character_dict_path is not None, "character_dict_path should not be None when character_type is ch"
+ def __init__(self, character_dict_path):
+ super(CTCLabelDecode, self).__init__()
+ self.character_str = []
+ assert character_dict_path is not None, "character_dict_path should not be None"
with open(character_dict_path, "rb") as fin:
lines = fin.readlines()
for line in lines:
line = line.decode('utf-8').strip("\n").strip("\r\n")
- self.character_str += line
- if use_space_char:
- self.character_str += " "
- dict_character = list(self.character_str)
+ self.character_str.append(line)
+ dict_character = self.character_str
- self.character_type = character_type
dict_character = self.add_special_char(dict_character)
self.dict = {}
for i, char in enumerate(dict_character):
self.dict[char] = i
self.character = dict_character
+ def __call__(self, preds, label=None, *args, **kwargs):
+ preds_idx = preds.argmax(axis=2)
+ preds_prob = preds.max(axis=2)
+ text = self.decode(preds_idx, preds_prob, is_remove_duplicate=True)
+ if label is None:
+ return text
+ label = self.decode(label)
+ return text, label
+
def add_special_char(self, dict_character):
+ dict_character = ['blank'] + dict_character
return dict_character
- def decode(self, text_index, text_prob=None, is_remove_duplicate=True):
+ def get_ignored_tokens(self):
+ return [0] # for ctc blank
+
+ def decode(self, text_index, text_prob=None, is_remove_duplicate=False):
""" convert text-index into text-label. """
result_list = []
ignored_tokens = self.get_ignored_tokens()
@@ -63,62 +103,3 @@ class BaseRecLabelDecode(object):
result_list.append((text, np.mean(conf_list)))
return result_list
- def get_ignored_tokens(self):
- return [0] # for ctc blank
-
-
-class CTCLabelDecode(BaseRecLabelDecode):
- """ Convert between text-label and text-index """
-
- def __init__(self,
- character_dict_path=None,
- character_type='ch',
- use_space_char=False,
- **kwargs):
- super(CTCLabelDecode, self).__init__(character_dict_path,
- character_type, use_space_char)
-
- def __call__(self, preds, label=None, *args, **kwargs):
- preds_idx = preds.argmax(axis=2)
- preds_prob = preds.max(axis=2)
- text = self.decode(preds_idx, preds_prob)
- if label is None:
- return text
- label = self.decode(label, is_remove_duplicate=False)
- return text, label
-
- def add_special_char(self, dict_character):
- dict_character = ['blank'] + dict_character
- return dict_character
-
-
-def get_image_file_list(img_file):
- imgs_lists = []
- if img_file is None or not os.path.exists(img_file):
- raise Exception("not found any img file in {}".format(img_file))
-
- img_end = {'jpg', 'bmp', 'png', 'jpeg', 'rgb', 'tif', 'tiff', 'gif', 'GIF'}
- if os.path.isfile(img_file) and imghdr.what(img_file) in img_end:
- imgs_lists.append(img_file)
- elif os.path.isdir(img_file):
- for single_file in os.listdir(img_file):
- file_path = os.path.join(img_file, single_file)
- if imghdr.what(file_path) in img_end:
- imgs_lists.append(file_path)
- if len(imgs_lists) == 0:
- raise Exception("not found any img file in {}".format(img_file))
- return imgs_lists
-
-
-def check_and_read_gif(img_path):
- if os.path.basename(img_path)[-3:] in ['gif', 'GIF']:
- gif = cv2.VideoCapture(img_path)
- ret, frame = gif.read()
- if not ret:
- print("Cannot read {}. This gif image maybe corrupted.")
- return None, False
- if len(frame.shape) == 2 or frame.shape[-1] == 1:
- frame = cv2.cvtColor(frame, cv2.COLOR_GRAY2RGB)
- imgvalue = frame[:, :, ::-1]
- return imgvalue, True
- return None, False
\ No newline at end of file
diff --git a/python/rapidOCR.sh b/python/rapidOCR.sh
index 9682f3e..d0148f1 100644
--- a/python/rapidOCR.sh
+++ b/python/rapidOCR.sh
@@ -1,4 +1,7 @@
-python RapidOCR.py --use_server False \
+#! /bin/bash
+set -e errexit
+
+python rapidOCR.py --use_server False \
--det_model_path models/ch_ppocr_mobile_v2.0_det_infer.onnx \
--cls_model_path models/ch_ppocr_mobile_v2.0_cls_infer.onnx \
--rec_model_path models/ch_ppocr_mobile_v2.0_rec_infer.onnx \