def post(self, **kwargs):
parser = reqparse.RequestParser()
parser.add_argument('ImgString', type=str,
required=True, location='form')
args = parser.parse_args()
rst = None
imgString = args['ImgString'].encode().split(b';base64,')[-1]
try:
b64_image = base64_to_PIL(imgString)
if b64_image is not None:
b64_image = np.array(b64_image)
result = text_predict(b64_image)
text = ' '.join([i['text'] for i in result])
rst = {
'code': "success",
'text': text
}
except ValueError:
rst = {
"code": "FAILURE",
"errMsg": "ocr访问出错",
"id": "1501"
}
return rst
def post(self, **kwargs):
parser = reqparse.RequestParser()
parser.add_argument('imgString', type=str, location='form')
args = parser.parse_args()
result = None
t = time.time()
uidJob = uuid.uuid1().__str__()
imgString = args['imgString'].encode().split(b';base64,')[-1]
b64_image = base64_to_PIL(imgString)
if b64_image:
b64_image = np.array(b64_image)
# 避免识别过程中重复上传识别
while time.time() - t <= TIMEOUT:
if os.path.exists(filelock):
continue
else:
with open(filelock, 'w') as f:
f.write(uidJob)
# 图片识别
result = text_predict(b64_image)
os.remove(filelock)
break
return jsonify({'text': [i['text'] for i in result]})
def ocr_post():
data = request.json
billModel = data.get('billModel', '')
# textAngle = data.get('textAngle',False)##文字检测
textLine = data.get('textLine', False) ##只进行单行识别
imgString = data['imgString'].encode().split(b';base64,')[-1]
img = base64_to_PIL(imgString)
if img is not None:
img = np.array(img)
res = run_ocr(img, billModel, textLine)
return json.dumps({'res': res}, ensure_ascii=False)
def get_job(self, callback):
##获取队列中的图像
ind = self.get_list(jobListTable)
if ind is not None:
value = self.get_dict(imgStringTable, [ind])
value = value[0]
if value is not None:
value = base64_to_PIL(value)
if value is not None:
value = value.convert("L")
value = callback(value)
else:
value = ""
print(ind, value)
self.set_dict(resSetTable, ind, value) ##存储识别结果到set
def POST(self):
t = time.time()
data = web.data()
uidJob = uuid.uuid1().__str__()
data = json.loads(data)
# print(data)
billModel = data.get('billModel','')
# textAngle = data.get('textAngle',False)##文字检测
textLine = data.get('textLine',False)##只进行单行识别
imgString = data['imgString'].encode().split(b';base64,')[-1]
img = base64_to_PIL(imgString)
if img is not None:
img = np.array(img)
H,W = img.shape[:2]
while time.time()-t<=TIMEOUT:
if os.path.exists(filelock):
continue
else:
with open(filelock,'w') as f:
f.write(uidJob)
if textLine:
##单行识别
partImg = Image.fromarray(img)
text = crnn_handle.predict(partImg)
res =[ {'text':text,'name':'0','box':[0,0,W,0,W,H,0,H]} ]
os.remove(filelock)
break
else:
# detectAngle = textAngle
result= text_predict(img)
if billModel=='' or billModel=='通用OCR' :
# result = union_rbox(result,0.2)
res = [{'text':x['text'],
'name':str(i),
'box':{'cx':x['cx'],
'cy':x['cy'],
'w':x['w'],
'h':x['h'],
'angle':x['degree']
}
} for i,x in enumerate(result)]
# res = adjust_box_to_origin(img,angle, res)##修正box
elif billModel=='火车票':
res = trainTicket.trainTicket(result)
res = res.res
res =[ {'text':res[key],'name':key,'box':{}} for key in res]
elif billModel=='身份证':
res = idcard.idcard(result)
res = res.res
res =[ {'text':res[key],'name':key,'box':{}} for key in res]
os.remove(filelock)
break
timeTake = time.time()-t
return json.dumps({'res':res,'timeTake':round(timeTake,4)},ensure_ascii=False)
def POST(self):
t = time.time()
data = web.data()
uidJob = uuid.uuid1().__str__()
data = json.loads(data)
billModel = data.get('billModel', '')
textAngle = data.get('textAngle', False) ##文字检测
textLine = data.get('textLine', False) ##只进行单行识别
imgString = data['imgString'].encode().split(b';base64,')[-1]
img = base64_to_PIL(imgString)
if img is not None:
img = np.array(img)
H, W = img.shape[:2]
while time.time() - t <= TIMEOUT:
if os.path.exists(filelock):
continue
else:
with open(filelock, 'w') as f:
f.write(uidJob)
if textLine:
##单行识别
partImg = Image.fromarray(img)
text = crnn.predict(partImg.convert('L'))
res = [{
'text': text,
'name': '0',
'box': [0, 0, W, 0, W, H, 0, H]
}]
os.remove(filelock)
break
else:
detectAngle = textAngle
result, angle = model.model(
img,
scale=scale,
maxScale=maxScale,
detectAngle=detectAngle, ##是否进行文字方向检测，通过web传参控制
MAX_HORIZONTAL_GAP=100, ##字符之间的最大间隔，用于文本行的合并
MIN_V_OVERLAPS=0.6,
MIN_SIZE_SIM=0.6,
TEXT_PROPOSALS_MIN_SCORE=0.1,
TEXT_PROPOSALS_NMS_THRESH=0.3,
TEXT_LINE_NMS_THRESH=0.99, ##文本行之间测iou值
LINE_MIN_SCORE=0.1,
leftAdjustAlph=0.01, ##对检测的文本行进行向左延伸
rightAdjustAlph=0.01, ##对检测的文本行进行向右延伸
)
if billModel == '' or billModel == '通用OCR':
result = union_rbox(result, 0.2)
res = [{
'text': x['text'],
'name': str(i),
'box': {
'cx': x['cx'],
'cy': x['cy'],
'w': x['w'],
'h': x['h'],
'angle': x['degree']
}
} for i, x in enumerate(result)]
res = adjust_box_to_origin(img, angle, res) ##修正box
elif billModel == '火车票':
res = trainTicket.trainTicket(result)
res = res.res
res = [{
'text': res[key],
'name': key,
'box': {}
} for key in res]
elif billModel == '身份证':
res = idcard.idcard(result)
res = res.res
res = [{
'text': res[key],
'name': key,
'box': {}
} for key in res]
os.remove(filelock)
break
timeTake = time.time() - t
return json.dumps({
'res': res,
'timeTake': round(timeTake, 4)
},
ensure_ascii=False)
def POST(self):
t = time.time()
data = web.data()
uidJob = uuid.uuid1().__str__()
data = json.loads(data)
# print(data)
billModel = data.get("billModel", "")
# textAngle = data.get('textAngle',False)##文字检测
textLine = data.get("textLine", False) ##只进行单行识别
imgString = data["imgString"].encode().split(b";base64,")[-1]
img = base64_to_PIL(imgString)
if img is not None:
img = np.array(img)
H, W = img.shape[:2]
while time.time() - t <= TIMEOUT:
if os.path.exists(filelock):
continue
else:
with open(filelock, "w") as f:
f.write(uidJob)
if textLine:
##单行识别
partImg = Image.fromarray(img)
text = crnn_handle.predict(partImg)
res = [{
"text": text,
"name": "0",
"box": [0, 0, W, 0, W, H, 0, H]
}]
os.remove(filelock)
break
else:
# detectAngle = textAngle
result = text_predict(img)
if billModel == "" or billModel == "通用OCR":
# result = union_rbox(result,0.2)
res = [{
"text": x["text"],
"name": str(i),
"box": {
"cx": x["cx"],
"cy": x["cy"],
"w": x["w"],
"h": x["h"],
"angle": x["degree"],
},
} for i, x in enumerate(result)]
# res = adjust_box_to_origin(img,angle, res)##修正box
elif billModel == "火车票":
res = trainTicket.trainTicket(result)
res = res.res
res = [{
"text": res[key],
"name": key,
"box": {}
} for key in res]
elif billModel == "身份证":
res = idcard.idcard(result)
res = res.res
res = [{
"text": res[key],
"name": key,
"box": {}
} for key in res]
os.remove(filelock)
break
timeTake = time.time() - t
return json.dumps({
"res": res,
"timeTake": round(timeTake, 4)
},
ensure_ascii=False)
def ocr():
if request.method == "POST":
request_start_time = time.time()
uid_job = uuid.uuid1().__str__()
data = request.json
# 模型参数
bill_model = data.get("billModel", "")
# 文字检测
# text_angle = data.get('textAngle', False)
# 只进行单行识别
text_line = data.get("textLine", False)
img_str = data["imgString"].encode().split(b";base64,")[-1]
img = base64_to_PIL(img_str)
if img is None:
response_time = time.time() - request_start_time
return jsonify({"res": [], "timeTake": round(response_time, 4)})
else:
img = np.array(img)
h, w = img.shape[:2]
final_result: list = []
while time.time() - request_start_time <= TIMEOUT:
if os.path.exists(file_lock):
continue
else:
with open(file_lock, "w") as f:
f.write(uid_job)
if text_line:
# 单行识别
part_img = Image.fromarray(img)
text = crnn_handle.predict(part_img)
final_result = [{
"text": text,
"name": "0",
"box": [0, 0, w, 0, w, h, 0, h]
}]
os.remove(file_lock)
break
else:
result = text_predict(img)
if bill_model == "" or bill_model == "通用OCR":
final_result = [{
"text": x["text"],
"name": str(i),
"box": {
"cx": x["cx"],
"cy": x["cy"],
"w": x["w"],
"h": x["h"],
"angle": x["degree"],
},
} for i, x in enumerate(result)]
elif bill_model == "火车票":
train_ticket_result = trainTicket.trainTicket(
result)
result = train_ticket_result.res
final_result = [{
"text": result[key],
"name": key,
"box": {}
} for key in result]
elif bill_model == "身份证":
id_card_result = idcard.idcard(result)
result = id_card_result.res
final_result = [{
"text": result[key],
"name": key,
"box": {}
} for key in result]
os.remove(file_lock)
break
response_time = time.time() - request_start_time
return jsonify({
"res": final_result,
"timeTake": round(response_time, 4)
})
def upload():
t = time.time()
parser = reqparse.RequestParser()
parser.add_argument('file')
args = parser.parse_args()
# data = request.json
uidJob = uuid.uuid1().__str__()
file = args['file']
# data = json.loads(data)
billModel = '通用OCR'
textAngle = True # 文字检测
imgString = file.encode().split(b';base64,')[-1]
img = base64_to_PIL(imgString)
if img is not None:
img = np.array(img)
else:
return 0
H, W = img.shape[:2]
while time.time() - t <= TIMEOUT:
if os.path.exists(filelock):
continue
else:
with open(filelock, 'w') as f:
f.write(uidJob)
detectAngle = textAngle
result, angle = model.model(img,
scale=scale,
maxScale=maxScale,
detectAngle=detectAngle, # 是否进行文字方向检测，通过web传参控制
MAX_HORIZONTAL_GAP=100, # 字符之间的最大间隔，用于文本行的合并
MIN_V_OVERLAPS=0.6,
MIN_SIZE_SIM=0.6,
TEXT_PROPOSALS_MIN_SCORE=0.1,
TEXT_PROPOSALS_NMS_THRESH=0.3,
TEXT_LINE_NMS_THRESH=0.99, # 文本行之间测iou值
LINE_MIN_SCORE=0.1,
leftAdjustAlph=0.01, # 对检测的文本行进行向左延伸
rightAdjustAlph=0.01, # 对检测的文本行进行向右延伸
)
if billModel == '' or billModel == '通用OCR':
result = union_rbox(result, 0.2)
res = {'name': '', 'position': '', 'company': '', 'local': '', 'email': '', 'phone': '',
'other': ''}
# 对公司进行提取
companyKeyword = ['公司', 'company', '银行', 'Bank', '集团', 'Group', '商行', 'Factory', 'CO', 'LTD']
for i, x in enumerate(result):
for ck in companyKeyword:
if ck in x['text']:
res['company'] = x['text']
result.pop(i)
break
if res['company']:
break
name = ''
name_height = 0
pop_data = 0
# 对姓名进行处理
for i, x in enumerate(result):
if 'name' in x['text'] or '姓名' in x['text']:
res['name'] = x['text'].replace('name', '')
res['name'] = res['name'].replace('姓名', '')
res['name'] = res['name'].replace(':', '')
result.pop(i)
break
else:
# 正则取中文，进行分别进行中文名和英文名的判断
text_list = re.findall(r'[\u4e00-\u9fa5]', x['text'])
if text_list:
if name_height < x['h'] and 1 < len(text_list) < 5:
name_height = x['h']
name = x['text']
pop_data = i
else:
strUpper = x['text'].upper()
if name_height < x['h'] and strUpper != x['text']:
name_height = x['h']
name = x['text']
pop_data = i
res['name'] = name
result.pop(pop_data)
# 对职位进行提取
pos = ['GM', 'VP', 'HRD', 'OD', 'MD', 'OM', 'PM', 'BM', 'DM', 'RM', 'AAD', 'ACD', 'AD', 'AE',
'AP', 'ASM', 'VC', 'CWO', 'COO', 'CXO', 'CYO', 'CZO', 'PS', 'Manager', 'Engineer']
posNum = "".join(filter(str.isdigit, x['text']))
for i, x in enumerate(result[pop_data:pop_data + 3]):
if '职位' in x['text'] or 'position' in x['text']:
res['position'] = x['text'].replace('职位', '')
res['position'] = res['position'].replace('position', '')
res['position'] = res['position'].replace(':', '')
result.pop(i)
break
if not res['position']:
for i, x in enumerate(result[pop_data:pop_data + 3]):
for ch in x['text']:
if u'\u4e00' <= ch <= u'\u9fff' and len(posNum) == 0:
res['position'] = x['text']
result.pop(i)
break
if res['position']:
break
for p in pos:
if p in x['text']:
res['position'] = x['text']
result.pop(i)
break
if res['position']:
break
# 对邮箱进行提取
for i, x in enumerate(result):
if '@' in x['text'] or '邮箱' in x['text'] or 'email' in x['text']:
res['email'] = x['text'].replace('邮箱', '')
res['email'] = res['email'].replace('email', '')
res['email'] = res['email'].replace(':', '')
result.pop(i)
break
# 对地址进行提取
localKeyword = ['市', '省', '区', '号', '路', '岛', '地址', 'sheng', 'shi', 'qu', 'hao', 'lu']
for i, x in enumerate(result):
for k in localKeyword:
if k in x['text'] or 'Add' in x['text'] or 'add' in x['text']:
res['local'] = x['text'].replace('地址', '')
res['local'] = res['local'].replace('Add', '')
res['local'] = res['local'].replace(':', '')
result.pop(i)
break
if res['local']:
break
# 对手机号码进行提取
for i, x in enumerate(result):
if '-' not in x['text'] and '-' not in x['text']:
telephone = "".join(filter(str.isdigit, x['text']))
if 11 <= len(telephone) <= 13:
res['phone'] += x['text'].replace('电话', '') + ';'
res['phone'] = res['phone'].replace('Tel', '')
res['phone'] = res['phone'].replace('手机', '')
res['phone'] = res['phone'].replace(':', '')
result.pop(i)
# 其他文本提取
for i, x in enumerate(result):
res['other'] += x['text'] + ';'
elif billModel == '火车票':
res = trainTicket.trainTicket(result)
res = res.res
res = [{'text': res[key], 'name': key, 'box': {}} for key in res]
elif billModel == '身份证':
res = idcard.idcard(result)
res = res.res
res = [{'text': res[key], 'name': key, 'box': {}} for key in res]
os.remove(filelock)
break
timeTake = time.time() - t
return jsonify({
"code": "0000",
"msg": "成功",
"data": res
})
def POST(self):
data = web.data()
data = json.loads(data)
CommandID = data.get('commandID', '')
BusinessID = data.get('businessID','')
SessionID = data.get('sessionID','')
# 下面三行兼容原有的web app demo
billModel = data.get('billModel','') ## 确定具体使用哪种模式识别
textAngle = data.get('textAngle', False) ## 文字方向检测
textLine = data.get('textLine', False) ## 只进行单行识别
# 处理传递参数
if CommandID != '':
# self.logger.info('post request from JiuTian IP= %s ,CommandID=%s' % (web.ctx.get('ip'), CommandID))
if CommandID == '100001':
billModel = 'invoice'
elif CommandID == '200001':
billModel = 'idcard'
elif CommandID == '300001':
billModel = 'bankcard'
elif CommandID == '400001':
billModel = 'licenseplate'
else:
## 返回请求参数错误
return json.dumps(
{'sessionID': SessionID,
'commandID': CommandID,
'businessID': BusinessID,
'timeStamp': time.strftime('%Y%m%d%H%M%S', time.localtime()),
'execStatus': {"statusCode": 0x800003, "statusDescription": "请求参数错误"},
'resultInfo': {}}, ensure_ascii=False
)
picName = data.get('picName', 'new.jpg')
picpath = 'http://172.31.201.35:18081' + data.get('picUrl', '') + picName
response = requests.get(picpath, stream=True)
## 处理可能出现的视频（只可能出现在‘licenseplate’中）
if picName.endswith(('.jpg', '.png', '.jpeg', '.JPG','.JPEG','.PNG')):
img = Image.open(BytesIO(response.content)).convert('RGB')
elif picName.endswith(('.mp4','.MP4','.avi','.AVI')) and billModel == 'licenseplate':
with open(picName, 'wb+') as f:
f.write(response.content)
saveName = picName.split('.')[0]+'_new.mp4'
result = model_lp.model_video(picName,saveName)
res = {'carNo': list(result), 'picUrl': '', 'picName': ''}
upload_url = 'http://172.31.201.35:18081' + '/cmcc-ocr-webapi-1.0/service/remoteUploadPic/'
files = {'image': (saveName, open(saveName, 'rb'), 'image/jpeg', {})}
reply = requests.post(upload_url, files=files)
# get the picUrl and picName
reply = reply.json()
# print(reply)
res['picUrl'] = reply['picUrl']
res['picName'] = reply['picName']
# delete tmp files
# os.remove(picName)
#os.remove(saveName)
return json.dumps({'sessionID': SessionID,
'commandID': CommandID,
'businessID': BusinessID,
'timeStamp': time.strftime('%Y%m%d%H%M%S', time.localtime()),
'execStatus': {"statusCode": 0x000000, "statusDescription": "成功"},
'resultInfo': res}, ensure_ascii=False)
else:
## 返回请求参数错误
return json.dumps(
{'sessionID': SessionID,
'commandID': CommandID,
'businessID': BusinessID,
'timeStamp': time.strftime('%Y%m%d%H%M%S', time.localtime()),
'execStatus': {"statusCode": 0x800004, "statusDescription": "内部数据错误"},
'resultInfo': {}}, ensure_ascii=False
)
else:
## 兼容原有的web app demo
imgString = data['imgString'].encode().split(b';base64,')[-1]
img = base64_to_PIL(imgString)
if img is not None:
img = np.array(img)
H, W = img.shape[:2]
timeTake = time.time()
if textLine:
##单行识别
partImg = Image.fromarray(img)
# text = model.crnnOcr(partImg.convert('L'))
text = crnn.predict(partImg.convert('L'))
res = [{'text': text, 'name': '0', 'box': [0, 0, W, 0, W, H, 0, H]}]
else:
if billModel == 'licenseplate':
img = Image.fromarray(img)
img, result = model_lp.model_MTCNN(img)
res = self.format_text(result, img, 0, billModel, CommandID)
else:
detectAngle = textAngle
result, angle = model.model(img,
scale=scale,
maxScale=maxScale,
detectAngle=detectAngle, ##是否进行文字方向检测，通过web传参控制
MAX_HORIZONTAL_GAP=30, ##字符之间的最大间隔，用于文本行的合并
MIN_V_OVERLAPS=0.6,
MIN_SIZE_SIM=0.6,
TEXT_PROPOSALS_MIN_SCORE=0.07,
TEXT_PROPOSALS_NMS_THRESH=0.3,
TEXT_LINE_NMS_THRESH=0.99, ##文本行之间测iou值
LINE_MIN_SCORE=0.07,
leftAdjustAlph=0.02, ##对检测的文本行进行向左延伸
rightAdjustAlph=0.02, ##对检测的文本行进行向右延伸
)
res = self.format_text(result, img, angle, billModel, CommandID)
timeTake = time.time() - timeTake
## 输出，同样区分是否是原有的web app demo接口
if CommandID == '':
# os.remove(path)
# print(res)
# outpic = self.plot_boxes(img, angle, result, color=(0, 0, 0))
# outpic.save('new.jpg')
return json.dumps({'res': res, 'timeTake': round(timeTake, 4)}, ensure_ascii=False)
else:
if timeTake > 15:
return json.dumps(
{'sessionID': SessionID,
'commandID': CommandID,
'businessID': BusinessID,
'timeStamp': time.strftime('%Y%m%d%H%M%S', time.localtime()),
'execStatus': {"statusCode": 0x800001, "statusDescription": "响应超时"},
'resultInfo': {}}, ensure_ascii=False
)
# save and upload the box pic
if billModel == 'licenseplate':
outpic = Image.fromarray(img)
else:
outpic = self.plot_boxes(img, angle, result, color=(0, 0, 0))
outpic.save(picName)
upload_url = 'http://172.31.201.35:18081' + '/cmcc-ocr-webapi-1.0/service/remoteUploadPic/'
files = {'image': (picName, open(picName, 'rb'), 'image/jpeg', {})}
reply = requests.post(upload_url, files=files)
# get the picUrl and picName
reply = reply.json()
# print(reply)
res['picUrl'] = reply['picUrl']
res['picName'] = reply['picName']
# delete tmp files
os.remove(picName)
return json.dumps({'sessionID': SessionID,
'commandID': CommandID,
'businessID': BusinessID,
'timeStamp': time.strftime('%Y%m%d%H%M%S', time.localtime()),
'execStatus': {"statusCode": 0x000000, "statusDescription": "成功"},
'resultInfo': res}, ensure_ascii=False)
