def img_word_translate(img_path):
ocr = PaddleOCR(use_angle_cls=True, lang="ch")
result = ocr.ocr(img_path, cls=True)
image = Image.open(img_path).convert('RGB')
boxes = [line[0] for line in result]
txts = [line[1][0] for line in result]
scores = [line[1][1] for line in result]
im_show = draw_ocr(image, boxes, txts, scores, font_path='/doc/simfang.ttf')
im_show = Image.fromarray(im_show)
n = 0
stR = ""
for line in result:
# print(line[1])
n += 1
texts = line[1][0]
if re.findall('[\u3002\uff1b\uff0c\uff1a\u201c\u201d\uff08\uff09\u3001\uff1f\u300a\u300b]', texts):
stR = stR + str(texts)
else:
stR = stR + str(texts) + ","
# print(texts)
print(stR)
translate(stR, "en", "zh-CN")
im_show.show()
return stR
def ocr_once():
ocr = PaddleOCR(use_angle_cls=True, lang="ch")
img_path = '/home/jiantang/桌面/tmp/certification/认证合一/归档/道路运输证/c1_001.png'
image = cv2.imread(img_path)
result = ocr.ocr(image, cls=True)
# 保存框框结果
boxes = [line[0] for line in result]
txts = [line[1][0] for line in result]
scores = [line[1][1] for line in result]
for i in range(len(boxes)):
pts = np.array(boxes[i], np.int32)
pts = pts.reshape((-1, 1, 2))
cv2.polylines(image, [pts], True, (0, 255, 255))
cv2.imwrite('tmp_img.png', image)
# 保存文字结果
img = np.zeros((image.shape[0], image.shape[1], 3), np.uint8)
for i in range(len(boxes)):
pts = np.array(boxes[i], np.int32)
pts = pts.reshape((-1, 1, 2))
cv2.polylines(img, [pts], True, (0, 255, 255))
img = cv2ImgAddText(img, txts[i], int(boxes[i][0][0] - 1),
int(boxes[i][0][1] - 1), (0, 255, 255), 20)
cv2.imwrite('tmp_txt.png', img)
cv2.imshow('line', img)
cv2.waitKey()
def ocr_once(path):
ocr = PaddleOCR(use_angle_cls=True, lang="ch")
img = cv2.imread(path)
result = ocr.ocr(img, cls=True)
# print(type(result))
print(result)
return check_c1(result)
def __init__(self, args):
self.args = args
self.max_seq_length = args.max_seq_length
# init ser token and model
tokenizer_class, base_model_class, model_class = MODELS[
args.ser_model_type]
self.tokenizer = tokenizer_class.from_pretrained(
args.model_name_or_path)
self.model = model_class.from_pretrained(args.model_name_or_path)
self.model.eval()
# init ocr_engine
from paddleocr import PaddleOCR
self.ocr_engine = PaddleOCR(
rec_model_dir=args.rec_model_dir,
det_model_dir=args.det_model_dir,
use_angle_cls=False,
show_log=False)
# init dict
label2id_map, self.id2label_map = get_bio_label_maps(
args.label_map_path)
self.label2id_map_for_draw = dict()
for key in label2id_map:
if key.startswith("I-"):
self.label2id_map_for_draw[key] = label2id_map["B" + key[1:]]
else:
self.label2id_map_for_draw[key] = label2id_map[key]
def one_pred(img_path):
ocr = PaddleOCR(use_angle_cls=True, lang="ch")
result = ocr.ocr(img_path, cls=True)
image = Image.open(img_path).convert('RGB')
boxes = [line[0] for line in result]
txts = [line[1][0] for line in result]
scores = [line[1][1] for line in result]
im_show = draw_ocr(image,
boxes,
txts,
scores,
font_path='/doc/simfang.ttf')
im_show = Image.fromarray(im_show)
n = 0
stR = ""
for line in result:
#print(line[1])
n += 1
texts = line[1][0]
stR = stR + "," + str(texts)
#print(texts)
print(stR)
im_show.show()
return stR
def GetText(path=None,result_pic=False):
"""
文字识别
:param path:图片路径
:return:
"""
if not path:
path = "F:/git/wxbot/dm/pic/222.png"
# ocr = PaddleOCR(use_gpu=False,det_model_dir="./inference/ch_det_mv3_db/",rec_model_dir="./inference/ch_det_mv3_db/") # need to run only once to download and load model into memory
ocr = PaddleOCR(gpu=False, det_model_dir="./inference/ch_det_r50_vd_db",
rec_model_dir="./inference/ch_rec_r34_vd_crnn_infer") # need to run only once to download and load model into memory
result = ocr.ocr(path)
txts = [line[1][0] for line in result]
# 展示图片
if result_pic:
image = Image.open(path).convert('RGB')
boxes = [line[0] for line in result]
scores = [line[1][1] for line in result]
im_show = draw_ocr(image, boxes, txts, scores)
im_show = Image.fromarray(im_show)
TMP_FILE = os.path.join(conf.TMP_PATH, "result.png")
im_show.save(TMP_FILE)
# print(txts)
return txts
def __init__(
self,
remove_numeric_tables: bool = False,
valid_languages: Optional[List[str]] = ["eng"],
):
"""
:param remove_numeric_tables: This option uses heuristics to remove numeric rows from the tables.
The tabular structures in documents might be noise for the reader model if it
does not have table parsing capability for finding answers. However, tables
may also have long strings that could possible candidate for searching answers.
The rows containing strings are thus retained in this option.
:param valid_languages: validate languages from a list of languages specified here
(https://tesseract-ocr.github.io/tessdoc/Data-Files-in-different-versions.html)
This option can be used to add test for encoding errors. If the extracted text is
not one of the valid languages, then it might likely be encoding error resulting
in garbled text. Run the following line of code to check available language packs:
# List of available languages
print(pytesseract.get_languages(config=''))
"""
# save init parameters to enable export of component config as YAML
self.set_config(remove_numeric_tables=remove_numeric_tables,
valid_languages=valid_languages)
self.recognize = PaddleOCR(use_angle_cls=True, lang='ch')
super().__init__(remove_numeric_tables=remove_numeric_tables,
valid_languages=valid_languages)
def convert_from_url(img_url):
ocr = PaddleOCR(use_angle_cls=True, lang="ch")
response = requests.get(img_url)
image = np.array(Image.open(BytesIO(response.content)))
result = ocr.ocr(image, cls=True)
text = [line[1][0] for line in result]
return " ".join(text)
def one_pred(img_path):
ocr = PaddleOCR(use_angle_cls=True, lang="ch")
result = ocr.ocr(img_path, cls=False)
# print(result)
n = 0
stR = ""
phone = "" #电话
shopName = "" #店铺名
describe = "" #描述
otherinfo = "没有其他信息"
print("+++", result)
for line in result:
n += 1
texts = line[1][0]
if texts == "":
continue
if len(texts) < 2:
continue
r = test.pre(texts)
if len(texts) < 3:
continue
if r == "商铺名":
shopName = shopName + texts
continue
s = ms.pre(texts)
if s == "描述":
describe = describe + texts
continue
stR = stR + str(line[1][0]) + ":" + str(line[1][1]) + "\n"
phones = texts.split("1")
# print("++",phone)
for i in phones:
# print(i)
if re.match(r'^[1-9]\d{9}$', i):
a = "1" + i
# print(a)
phone = phone + ";" + a
else:
phone = " 图片中没有电话信息"
phone = phone[1:]
print("****", shopName, phone, describe, otherinfo)
#return im_show.show()
image = Image.open(img_path).convert('RGB')
boxes = [line[0] for line in result]
txts = [line[1][0] for line in result]
scores = [line[1][1] for line in result]
im_show = draw_ocr(image,
boxes,
txts,
scores,
font_path='/doc/simfang.ttf')
im_show = Image.fromarray(im_show)
# im_show.show()
# im_show.save是保存识别后的图片
fanhui_img = 'result.jpg'
im_show.save(fanhui_img)
return shopName, phone, describe, otherinfo, fanhui_img
def __init__(self):
self.ocr_model = PaddleOCR(
use_angle_cls=True,
# det='ch_ppocr_server_v1.1_det',
det_model_dir='./pretrained_models/ch_ppocr_server_v2.0_det_infer',
# rec='ch_ppocr_server_v1.1_rec',
rec_model_dir='./pretrained_models/ch_ppocr_server_v2.0_rec_infer',
cls='ch_ppocr_mobile_v1.1_cls')
def init():
ocr = PaddleOCR(
use_angle_cls=True, lang="ch"
) # need to run only once to download and load model into memory
img_path = './data/3.jpg'
result = ocr.ocr(img_path, cls=True)
for line in result:
print(line)
def __init__(self):
self.model = PaddleOCR(use_angle_cls=True, lang="ch")
with open(os.path.join(os.getcwd(), "confs/certificate.yaml"),
"r",
encoding='UTF-8') as f:
self.config = yaml.load(f, Loader=yaml.FullLoader)
self.tmp_result = {}
self.pos_dict = {} # To tell client witch certificate the image is.
self.face_feature = []
self.reset()
def ocr_run():
global ocr_init
ocr = PaddleOCR(use_angle_cls=True, use_gpu=True, use_tensorrt=False, gpu_mem=3000, det_max_side_len=960, rec_batch_num=2)
cnt = 0
last_result = []
while not b_exit:
try:
frame = ocr_queue.get(block=True, timeout=0.1)
except:
continue
print('ocr got one frame ....')
t0 = time.time()
result = ocr.ocr(frame, cls=True)
t1 = time.time()
#try:
# speech_que.put_nowait(result)
#except:
# pass
if len(result) < 1:
continue
try:
if last_result:
likehood = samelike(last_result, result)
if likehood >= 0.6: # and len(last_result) == len(result):
Speech.Block_Speech_text("请翻页")
last_result = result
continue
last_result = result
except:
traceback.print_exc()
pass
if 1:
text_lst = []
for line in result:
text = line[1][0]
text_lst.append(text)
print(line)
for text in text_lst:
Speech.Block_Speech_text(text)
cnt += 1
if cnt % 1 == 0:
print("frame cnt [%d] ocr detect delay = %.1fms" % (cnt, (t1 - t0) * 1000))
def convertText2(self, img):
ocr = PaddleOCR(
) # need to run only once to download and load model into memory
result = ocr.ocr(
img) # return a list with coordinates, content and accuracy
index = 0
done = False
for line in result:
if done: break
[corList, (txt, _)] = line
[[_, y1], [x2, _], [_, _], [_, y4]] = corList
if '期货品种' in txt:
self.findContent2(index, y1, y4, x2, result, '期货品种')
elif '现货头寸' in txt:
self.findContent2(index, y1, y4, x2, result, '现货头寸')
elif '期货头寸' in txt:
self.findContent2(index, y1, y4, x2, result, '期货头寸')
elif '现货数量' in txt:
self.findContent2(index, y1, y4, x2, result, '现货数量')
elif '手' in txt:
self.findContent2(index, y1, y4, x2, result, '合约数量')
elif '计划套保' in txt:
[corListPrev, (txtP, _)] = result[index - 1]
[corListAft, (txtA, _)] = result[index + 1]
[[_, yp1], _, _, _] = corListPrev
[[_, ya1], _, _, _] = corListAft
if abs(yp1 - y1) < abs(ya1 - y1):
self.textDict['hedgeRatio'] = txtP
else:
self.textDict['hedgeRatio'] = txtA
elif '保证金规模' in txt:
self.findContent2(index, y1, y4, x2, result, '保证金规模')
elif txt == '最大亏损限额':
self.findContent2(index, y1, y4, x2, result, '最大亏损')
elif txt == '行情分析':
self.get_multi_lines(index, result, '行情分析')
elif txt == '持仓预案':
self.get_multi_lines(index, result, '持仓预案')
elif txt == '平仓预案':
self.get_multi_lines(index, result, '平仓预案')
elif '止盈位' in txt:
self.get_multi_lines(index, result, '止盈')
elif '止损位' in txt:
self.get_multi_lines(index, result, '止损')
elif '备注' in txt:
self.get_multi_lines(index, result, '备注')
done = True
index += 1
def demo_use():
# Paddleocr目前支持中英文、英文、法语、德语、韩语、日语，可以通过修改lang参数进行切换，参数依次为`ch`, `en`, `french`, `german`, `korean`, `japan`。
ocr = PaddleOCR(use_angle_cls=True, lang="en", use_gpu=False)
# 输入待识别图片路径
img_path = "C:/Users/FH/Desktop/ocrtupianjieya/icpr_mtwi_task1/test_line_image_transfer/line_101415.jpg"
# img_path = "C:/Users/FH/Desktop/line_101415_angel.jpg"
# img_path = "../pillow_toolkits/line_100031_higher_gray.jpg"
# 输出结果保存路径
result = ocr.ocr(img_path, det=False, cls=True)
print(result)
def one_pred(img_path):
ocr = PaddleOCR(use_angle_cls=True, lang="ch")
result = ocr.ocr(img_path, cls=True)
image = Image.open(img_path).convert('RGB')
boxes = [line[0] for line in result]
txts = [line[1][0] for line in result]
scores = [line[1][1] for line in result]
im_show = draw_ocr(image,
boxes,
txts,
scores,
font_path='/doc/simfang.ttf')
im_show = Image.fromarray(im_show)
im_show.show()
def ocr_preprocess(img_dir):
ocr = PaddleOCR(use_angle_cls=True, lang="ch", use_gpu=True)
ocr_reses = []
img_names = sorted(
os.listdir(img_dir), key=lambda x: int(x.split("_")[1].split(".")[0]))
for img_name in img_names:
img_path = os.path.join(img_dir, img_name)
parsing_res = ocr.ocr(img_path, cls=True)
ocr_res = []
for para in parsing_res:
ocr_res.append({"text": para[1][0], "bbox": para[0]})
ocr_reses.append((img_name, ocr_res))
return ocr_reses
def create_ocr(self, lang, use_gpu=False):
try:
ocr = PaddleOCR(lang=lang, use_gpu=use_gpu or self.use_gpu, use_angle_cls=True, gpu_mem=200)
except AssertionError:
print('创建OCR实例失败: lang={}'.format(lang))
else:
self.func[lang] = ocr
def __init__(self):
# self.ocr = PaddleOCR(use_angle_cls=True)
self.ocr = {
# "en":PaddleOCR(lang="en"),
"ch": PaddleOCR(lang="ch")
}
self.drop_scores = 0.5
def __init__(self):
dynaconf_config_ocr = settings.get('ocr', {})
det_model_dir = os.getcwd() + dynaconf_config_ocr.det_model_dir
rec_model_dir = os.getcwd() + dynaconf_config_ocr.rec_model_dir
rec_char_dict_path = os.getcwd(
) + dynaconf_config_ocr.rec_char_dict_path
cls_model_dir = os.getcwd() + dynaconf_config_ocr.cls_model_dir
self.min_angle = dynaconf_config_ocr.min_angle
self.max_angle = dynaconf_config_ocr.max_angle
self.angle_step = dynaconf_config_ocr.angle_step
self.text_score_thresh = dynaconf_config_ocr.text_score_thresh
self.ocr = PaddleOCR(
det_model_dir=det_model_dir,
rec_model_dir=rec_model_dir,
rec_char_dict_path=rec_char_dict_path,
cls_model_dir=cls_model_dir,
use_angle_cls=dynaconf_config_ocr.angle_switch,
use_gpu=False,
)
def detect_OCR(self, img):
# Paddleocr目前支持中英文、英文、法语、德语、韩语、日语，可以通过修改lang参数进行切换
# 参数依次为`ch`, `en`, `french`, `german`, `korean`, `japan`。
ocr = PaddleOCR(use_angle_cls=True, use_gpu=False, lang="ch", use_space_char=True,det_db_unclip_ratio=2.0) # det_db_unclip_ratio=2.5这是参数定义位置，这个参数是检测后处理时控制文本框大小的，默认2.0，可以尝试改成2.5或者更大，反之，如果觉得文本框不够紧凑，也可以把该参数调小。 need to run only once to download and load model into memory
img_path = img
result = ocr.ocr(img_path,cls=False)
for line in result:
print(line)
# 显示结果
from PIL import Image
image = Image.open(img_path).convert('RGB')
boxes = [line[0] for line in result]
txts = [line[1][0] for line in result]
scores = [line[1][1] for line in result]
im_show = draw_ocr(image, boxes, txts, scores, font_path='/path/to/PaddleOCR/doc/simfang.ttf')
im_show = Image.fromarray(im_show)
im_show.save('result.jpg')
return boxes, txts
def demo_use2():
ocr = PaddleOCR(use_angle_cls=True, lang="ch")
file_dir = "C:/Users/FH/Desktop/ocrtupianjieya/icpr_mtwi_task1/test_line_image_transfer"
file_names = os.listdir(file_dir)
reader = open(
"C:/Users/FH/Desktop/ocrtupianjieya/icpr_mtwi_task1/result_transfer.txt",
'r',
encoding='utf8')
already_files = [i.strip().split(" ")[0] for i in reader.readlines()]
reader.close()
writer = open(
"C:/Users/FH/Desktop/ocrtupianjieya/icpr_mtwi_task1/result_transfer.txt",
'a+',
encoding='utf8')
import time
begin_time = time.time()
count = 0
for index, file_name in enumerate(file_names):
if file_name in already_files:
continue
count += 1
out = ""
try:
file_path = os.path.join(file_dir, file_name)
result = ocr.ocr(file_path, det=False, cls=False)
if result:
out = result[0][0]
writer.write(file_name + " " + out + "\n")
if count <= 10:
print(result)
if (index + 1) % 500 == 0:
print("当前已处理{}条数据，目前处理平均速度为{}条/s".format(
index + 1, count / (time.time() - begin_time)))
writer.flush()
# if index == 5000:
# break
except Exception as e:
print("============error===========")
print(e)
print(file_name)
writer.write(file_name + " " + out + "\n")
writer.close()
def text_detection(input_file='../data/input/30800.jpg',
output_file='../data/output',
show=False,
method='google',
paddle_model=None):
'''
:param method: google or paddle
:param paddle_model: the preload paddle model for paddle ocr
'''
start = time.clock()
name = input_file.split('/')[-1][:-4]
ocr_root = pjoin(output_file, 'ocr')
img = cv2.imread(input_file)
if method == 'google':
print('*** Detect Text through Google OCR ***')
ocr_result = ocr.ocr_detection_google(input_file)
texts = text_cvt_orc_format(ocr_result)
texts = merge_intersected_texts(texts)
texts = text_filter_noise(texts)
texts = text_sentences_recognition(texts)
elif method == 'paddle':
# The import of the paddle ocr can be separate to the beginning of the program if you decide to use this method
from paddleocr import PaddleOCR
print('*** Detect Text through Paddle OCR ***')
if paddle_model is None:
paddle_model = PaddleOCR(use_angle_cls=True, lang="ch")
result = paddle_model.ocr(input_file, cls=True)
texts = text_cvt_orc_format_paddle(result)
else:
raise ValueError('Method has to be "google" or "paddle"')
visualize_texts(img,
texts,
shown_resize_height=800,
show=show,
write_path=pjoin(ocr_root, name + '.png'))
save_detection_json(pjoin(ocr_root, name + '.json'), texts, img.shape)
print("[Text Detection Completed in %.3f s] Input: %s Output: %s" %
(time.clock() - start, input_file, pjoin(ocr_root, name + '.json')))
def ocr_once():
# Paddleocr目前支持中英文、英文、法语、德语、韩语、日语，可以通过修改lang参数进行切换
# 参数依次为`ch`, `en`, `french`, `german`, `korean`, `japan`。
ocr = PaddleOCR(
use_angle_cls=True, lang="ch"
) # need to run only once to download and load model into memory
img_path = 'test/3.jpg'
img = cv2.imread(img_path)
start = time.time()
result = ocr.ocr(img, cls=True)
print(result)
print(time.time() - start)
# 显示结果
from PIL import Image
image = Image.open(img_path).convert('RGB')
boxes = [line[0] for line in result]
txts = [line[1][0] for line in result]
scores = [line[1][1] for line in result]
im_show = draw_ocr(image, boxes, txts, scores, font_path='simfang.ttf')
im_show = Image.fromarray(im_show)
im_show.save('result/result3.jpg')
def convertText1(self, img):
ocr = PaddleOCR(
) # need to run only once to download and load model into memory
result = ocr.ocr(
img) # return a list with coordinates, content and accuracy
index = 0
done = False
for line in result:
if done: break
[corList, (txt, _)] = line
[[x1, y1], [x2, y2], [x3, y3], [x4, y4]] = corList
if txt == '委托主体名称':
[corListPrev, (txtP, _)] = result[index - 1]
[corListAft, (txtA, _)] = result[index + 1]
[[_, yp1], _, _, _] = corListPrev
[[_, ya1], _, _, _] = corListAft
if abs(yp1 - y1) < abs(ya1 - y1):
self.textDict['entrustSubject'] = txtP
else:
self.textDict['entrustSubject'] = txtA
elif txt == '操作主体名称':
[corListPrev, (txtP, _)] = result[index - 1]
[corListAft, (txtA, _)] = result[index + 1]
[[_, yp1], _, _, _] = corListPrev
[[_, ya1], _, _, _] = corListAft
if abs(yp1 - y1) < abs(ya1 - y1):
self.textDict['operatorName'] = txtP
else:
self.textDict['operatorName'] = txtA
elif '现货品种' in txt:
self.findContent1(index, y1, y4, x2, result, '现货品种')
elif '开仓价格' in txt:
self.findContent1(index, y1, y4, x2, result, '开仓价格')
done = True
index += 1
class Vehicle_License_Plate_Recognition(nn.Layer):
def __init__(self):
super(Vehicle_License_Plate_Recognition, self).__init__()
self.vlpr = PaddleOCR(
det_model_dir=os.path.join(self.directory, 'det_vlpr'),
rec_model_dir=os.path.join(self.directory,
'ch_ppocr_server_v2.0_rec_infer'))
@staticmethod
def base64_to_cv2(b64str):
data = base64.b64decode(b64str.encode('utf8'))
data = np.frombuffer(data, np.uint8)
data = cv2.imdecode(data, cv2.IMREAD_COLOR)
return data
def plate_recognition(self, images=None):
assert isinstance(images, (list, str, np.ndarray))
results = []
if isinstance(images, list):
for item in images:
for bbox, text in self.vlpr.ocr(item):
results.append({'license': text[0], 'bbox': bbox})
elif isinstance(images, (str, np.ndarray)):
for bbox, text in self.vlpr.ocr(images):
results.append({'license': text[0], 'bbox': bbox})
return results
@serving
def serving_method(self, images):
if isinstance(images, list):
images_decode = [self.base64_to_cv2(image) for image in images]
elif isinstance(images, str):
images_decode = self.base64_to_cv2(images)
return self.plate_recognition(images_decode)
def ocr():
upload = request.files.get('upload')
lang = request.forms.get('lang')
print(lang)
global current_lang
global ocr
if current_lang!=lang:
current_lang=lang
ocr = PaddleOCR(lang=current_lang)
name, ext = os.path.splitext(upload.filename)
print(ext.lower())
if ext.lower() not in ('.png','.jpg','.jpeg'):
return "File extension not allowed."
timestamp=str(int(time.time()*1000))
savedName=timestamp+ext
save_path = "./uploaded/"
if not os.path.exists(save_path):
os.makedirs(save_path)
file_path = "{path}/{file}".format(path=save_path, file=savedName)
if os.path.exists(file_path)==True:
os.remove(file_path)
upload.save(file_path)
ret = {}
result = ocr.ocr(file_path)
text_lines=[]
for line in result:
text_line={}
index=0
for coord in line[0]:
text_line["x"+str(index)]=int(coord[0])
text_line["y"+str(index)]=int(coord[1])
index=index+1
text_line["text"]=line[1][0]
text_lines.append(text_line)
os.remove(file_path)
ret["text_lines"]=text_lines
return ret
def detect(image):
# PATH_IMG_IN = './in'
# filename = os.path.join(PATH_IMG_IN, '1.png')
# filename = 'FSRCNNp2x2.jpg'
# image = cv2.imread("picture2.jpg")
# cv2.imshow("a", image)
image1 = copy.deepcopy(image[0:int(image.shape[0]/2), 0:int(image.shape[1]/2), :])
# print(image.shape)
# cv2.imshow("b", image1)
# cv2.waitKey()
ocr = PaddleOCR() # need to run only once to download and load model into memory
start = time.perf_counter()
result = ocr.ocr(image1, rec=True)
# 第一个坐标表示的是第几个识别的字，第二个表示的识别的字中的是字的坐标还是还是字的内容或者置信度，第三个坐标表示的是四个坐
# 标中的第几个坐标，第四个坐标是是一个具体的坐标的x或者y
# print(result)
end = time.perf_counter()
print('检测文字区域 耗时{}'.format(end - start))
# 每个矩形，从左上角顺时针排列
for rect1 in result:
if 'XODTC' in rect1[1][0]:
# print(rect1[1][0])
return rect1[1][0]
def __init__(self, hwnd):
self.hwnd = hwnd
self.get_parent_size()
currfiletpath = Path(os.path.abspath(__file__))
self.currtpath = currfiletpath.parent
self.mous = PyMouse()
self.ditu_size = {
"jianye": (287, 142),
"donghaiwan": (119, 119),
"changancheng": (548, 277),
"jiangnanyewai": (159, 119),
"aolaiguo": (222, 150),
"huaguoshan": (159, 119),
"beijuluzhou": (226, 169),
"changshoujiaowai": (191, 167)
}
self.yunbiao_par = ["牛魔王", "观音姐姐", "镇元大仙", "孙婆婆", "地藏王"]
self.padocr = PaddleOCR(use_angle_cls=True, lang="ch")
# 调用百度的识别文字
APP_ID = '25963522'
API_KEY = 'RIOFdoDKGXfhHr2uLYVmUG8w'
SECRET_KEY = 'vnWA7s1WfWsrPwEvkgkiVI7MMxNTjFvw'
self.client = AipOcr(APP_ID, API_KEY, SECRET_KEY)
def gen_ocr_data(cls, imfile=None, **kwargs):
"""" 支持调用PaddleOCR进行预识别 """
from paddleocr import PaddleOCR
ppocr = PaddleOCR.get_paddleocr()
data = cls.gen_data(imfile, **kwargs)
lines = ppocr.ocr(str(imfile))
for line in lines:
pts, [text, score] = line
pts = [[int(p[0]), int(p[1])] for p in pts] # 转整数
sp = cls.gen_shape({
'text': text,
'score': round(float(score), 4)
}, pts)
data['shapes'].append(sp)
return data
