def My_OCR(url):
SUBSCRIPTION_KEY = os.getenv('OCR_SUBSCRIPTION_KEY')
ENDPOINT = os.getenv('OCR_ENDPOINT')
CV_CLIENT = ComputerVisionClient(
ENDPOINT, CognitiveServicesCredentials(SUBSCRIPTION_KEY))
ocr_results = CV_CLIENT.read(url, raw=True)
operation_location_remote = ocr_results.headers["Operation-Location"]
operation_id = operation_location_remote.split("/")[-1]
status = ["notStarted", "running"]
while True:
get_handw_text_results = CV_CLIENT.get_read_result(operation_id)
if get_handw_text_results.status not in status:
break
time.sleep(1)
succeeded = OperationStatusCodes.succeeded
text = []
if get_handw_text_results.status == succeeded:
res = get_handw_text_results.analyze_result.read_results
for text_result in res:
for line in text_result.lines:
if len(line.text) <= 8:
text.append(line.text)
r = re.compile("[0-9A-Z]{2,4}[.-]{1}[0-9A-Z]{2,4}")
text = list(filter(r.match, text))
return text[0].replace('.', '-') if len(text) > 0 else ""
class cos5year_vision:
computervision_client = None
##以下クラス内呼び出し専用
def __enter__(self):
self.computervision_client = ComputerVisionClient(endpoint, CognitiveServicesCredentials(subscription_key))
return self
def __exit__(self, exc_type, exc_value, traceback):
self.computervision_client = None
def __image_read(self,image_url):
recognize_results = self.computervision_client.read(image_url, raw=True)
operation_location_remote = recognize_results.headers["Operation-Location"]
operation_id = operation_location_remote.split("/")[-1]
return operation_id
def __get_read_result(self,operation_id):
##ここの無限ループどうにかする
while True:
get_text_results = computervision_client.get_read_result(operation_id)
if get_text_results.status not in ['notStarted', 'running']:
break
time.sleep(1)
if get_text_results.status == OperationStatusCodes.succeeded:
return get_text_results.analyze_result.read_results
##以下外部呼出し関数
def DetectTexts(self,image_url):
id = self.__image_read(image_url)
res = self.__get_read_result(id)
return [[{'text':line.text,'box':line.bounding_box} for line in text_result.lines] for text_result in res]
def read_image_text(self, client: ComputerVisionClient, remote_url: str):
"""
read text from an image using the Azure OCR Read API
"""
print(f"Calling read API on {remote_url}")
read_result = client.read(remote_url, raw=True)
operation_location_remote = read_result.headers['Operation-Location']
operation_id = operation_location_remote.split('/')[-1]
# GET method for read results
while True:
read_operation_result = client.get_read_result(operation_id)
if read_operation_result.status not in [
OperationStatusCodes.not_started,
OperationStatusCodes.running]:
break
time.sleep(1) # Re-check every second
return read_operation_result if read_operation_result.status ==\
OperationStatusCodes.succeeded else None
def ocr1(img):
#Cognitive Services endpoint and key
cog_key = '<Your Primary Key here>' #Paste your primary key here
cog_endpoint = '<Endpoint url here>' #Paste your endpoint here
# Get a client for the computer vision service
computervision_client = ComputerVisionClient(
cog_endpoint, CognitiveServicesCredentials(cog_key))
# Submit a request to read printed text in the image and get the operation ID
try:
recognize_handw_results = computervision_client.read(img, raw=True)
except Exception as e:
return str(e)
operation_location_remote = recognize_handw_results.headers[
"Operation-Location"]
operation_id = operation_location_remote.split("/")[-1]
while True:
get_handw_text_results = computervision_client.get_read_result(
operation_id)
if get_handw_text_results.status not in ['notStarted', 'running']:
break
time.sleep(1)
if get_handw_text_results.status == OperationStatusCodes.succeeded:
res = []
for text_result in get_handw_text_results.analyze_result.read_results:
for line in text_result.lines:
res.append(str(line.text))
if res != []:
res = str(" ".join(res)) #Result of OCR
else:
res = None
return res
def InvAns(strUrl):
if 'COMPUTER_VISION_SUBSCRIPTION_KEY' in os.environ:
subscription_key = os.environ['COMPUTER_VISION_SUBSCRIPTION_KEY']
else:
print("\nSet the COMPUTER_VISION_SUBSCRIPTION_KEY environment variable.\n**Restart your shell or IDE for changes to take effect.**")
sys.exit()
if 'COMPUTER_VISION_ENDPOINT' in os.environ:
endpoint = os.environ['COMPUTER_VISION_ENDPOINT']
else:
print("\nSet the COMPUTER_VISION_ENDPOINT environment variable.\n**Restart your shell or IDE for changes to take effect.**")
sys.exit()
computervision_client = ComputerVisionClient(endpoint, CognitiveServicesCredentials(subscription_key))
# print("===== Batch Read File - remote =====")
remote_image_handw_text_url = strUrl
recognize_handw_results = computervision_client.read(remote_image_handw_text_url, raw=True)
operation_location_remote = recognize_handw_results.headers["Operation-Location"]
operation_id = operation_location_remote.split("/")[-1]
while True:
get_handw_text_results = computervision_client.get_read_result(operation_id)
if get_handw_text_results.status not in ['notStarted', 'running']:
break
time.sleep(1)
d = {}
abcA = []
defgA = []
hijA = []
klmA = []
nopA= []
qrsA = []
uvwA = []
flat_list = []
if get_handw_text_results.status == OperationStatusCodes.succeeded:
for text_result in get_handw_text_results.analyze_result.read_results:
for sourceTxt in text_result.lines:
new_d = {}
new_d[sourceTxt.text] = sourceTxt.bounding_box
flat_list.append(sourceTxt.text)
def Get_shipping_in_pdf(sourceTxt):
# print(cf)
shp = r"Shipping.*\d.*$"
shp1 = r"Shipping"
if (re.findall(shp, sourceTxt)):
ans = re.findall(shp, sourceTxt)[0]
f_ans_split = ans.split()
for i in f_ans_split:
if i.isnumeric():
print(i)
f_ans = i
else:
f_ans = ''
elif (re.findall(shp1, sourceTxt)):
f_ans = re.findall(shp1, sourceTxt)[0]
else:
f_ans = "0"
return f_ans
def Get_company_in_pdf(sourceTxt):
frm = r"From:"
frm1 = r"From"
frm2= r"Invoice"
if(re.findall(frm, sourceTxt)):
f_ans = re.findall(frm, sourceTxt)[0]
elif(re.findall(frm1, sourceTxt)):
f_ans = re.findall(frm1, sourceTxt)[0]
elif(re.findall(frm2, sourceTxt)):
f_ans = re.findall(frm2, sourceTxt)[0]
else:
f_ans = 0
# print(f_ans)
return f_ans
def Get_order_in_pdf(sourceTxt):
prcnt = r"Purchase.*\d.*$"
prcnt1 = r"Ord.*\d.*$"
prcnt2 = r"Ord.*\D"
if (re.findall(prcnt, sourceTxt)):
ans = re.findall(prcnt, sourceTxt)[0]
f_ans_split = ans.split()
for i in f_ans_split:
if i.isnumeric():
f_ans = i
else:
f_ans = ''
elif(re.findall(prcnt1, sourceTxt)):
ans = re.findall(prcnt1, sourceTxt)[0]
f_ans_split = ans.split()
for i in f_ans_split:
if i.isnumeric():
f_ans = i
else:
f_ans = ''
elif (re.findall(prcnt2, sourceTxt)):
f_ans = re.findall(prcnt2, sourceTxt)[0]
else:
f_ans = ''
return f_ans
def Get_invoice_in_pdf(sourceTxt):
prcnt = r"Invoice.*\d.*$"
prcnt1 = r"INV.*\-\d.*$"
f_ans = ''
if (re.findall(prcnt1, sourceTxt)):
f_ans = re.findall(prcnt1, sourceTxt)[0]
elif(re.findall(prcnt, sourceTxt)):
ans = re.findall(prcnt, sourceTxt)[0]
f_ans_split = ans.split()
for i in f_ans_split:
if i.isnumeric():
f_ans = i
else:
f_ans = ''
else:
f_ans = ''
return f_ans
def Get_vat_in_document(sourceTxt):
prcnt = r"([0-9][0-9]%$|[0-9]%$)"
if(re.findall(prcnt, sourceTxt)):
f_ans = re.findall(prcnt, sourceTxt)[0]
else:
f_ans = 0
return f_ans
def Get_total_amount_in_pdf(sourceTxt):
lst = r"\d.*,\d{3}\.?\d*$"
dcm = r"\d.*\..?\d*$"
spc = r"\d.*\s\d{3}\.\d*$"
ans=[]
ttl = []
f_ans = []
if(re.findall(lst, sourceTxt)):
ans.extend(re.findall(lst, sourceTxt))
elif re.findall(dcm, sourceTxt):
ans.extend(re.findall(dcm, sourceTxt))
elif re.findall(spc, sourceTxt):
ans.extend(re.findall(spc, sourceTxt))
if ans:
for i in ans:
rem_c = i.split(',')
try:
ttl.append(float("".join(rem_c)))
except ValueError:
pass
if ttl == []:
pass
else:
f_ans.extend(ttl)
return f_ans
def Get_date_in_pdf(sourceTxt):
date_format = [ r"\d{2} (?:%s) \d{4}" % '|'.join(calendar.month_abbr[1:]), r"\d{1} (?:%s) \d{4}" % '|'.join(calendar.month_abbr[1:]), r"\d{2} (?:%s) \d{4}" % '|'.join(calendar.month_name), r"\d{1} (?:%s) \d{4}" % '|'.join(calendar.month_name[1:]), r"(?:%s) \d{2} \d{4}" % '|'.join(calendar.month_abbr[1:]), r"(?:%s) \d{1} \d{4}" % '|'.join(calendar.month_abbr[1:]), r"(?:%s) \d{2} \d{4}" % '|'.join(calendar.month_name), r"(?:%s) \d{1} \d{4}" % '|'.join(calendar.month_name[1:]),r"\d{2} (?:%s), \d{4}" % '|'.join(calendar.month_abbr[1:]), r"\d{1} (?:%s), \d{4}" % '|'.join(calendar.month_abbr[1:]), r"\d{2} (?:%s), \d{4}" % '|'.join(calendar.month_name), r"\d{1} (?:%s), \d{4}" % '|'.join(calendar.month_name[1:]), r"(?:%s) \d{2}, \d{4}" % '|'.join(calendar.month_abbr[1:]), r"(?:%s) \d{1}, \d{4}" % '|'.join(calendar.month_abbr[1:]), r"(?:%s) \d{2}, \d{4}" % '|'.join(calendar.month_name), r"(?:%s) \d{1}, \d{4}" % '|'.join(calendar.month_name[1:])]
dats = []
f_dats = ''
for i in date_format:
if(re.findall(i, sourceTxt)):
dats.extend(re.findall(i, sourceTxt))
else:
ans = re.findall(r'([1-9]|1[0-9]|2[0-9]|3[0-1]|0[0-9])(.|-|\/)([1-9]|1[0-2]|2[0-9]|3[0-9])(.|-|\/)(20[0-9][0-9])',sourceTxt)
ans_f = [''.join(ans[i]) for i in range(len(ans))]
dats.extend(ans_f)
for sublist in dats:
if any(sublist):
f_dats = sublist
break
return f_dats
abc = Get_date_in_pdf(sourceTxt.text)
defg = Get_total_amount_in_pdf(sourceTxt.text)
hij = Get_vat_in_document(sourceTxt.text)
klm = Get_invoice_in_pdf(sourceTxt.text)
nop = Get_order_in_pdf(sourceTxt.text)
qrs = Get_company_in_pdf(sourceTxt.text)
uvw = Get_shipping_in_pdf(sourceTxt.text)
if abc=="" and defg=="" and hij:
continue
elif abc:
abcA.append(abc)
continue
elif defg:
defgA.append(defg)
continue
elif hij:
hijA.append(hij)
elif klm:
klmA.append(klm)
elif nop:
nopA.append(nop)
elif qrs:
print(qrs)
qrsA.append(qrs)
elif uvw:
uvwA.append(uvw)
def get_index_ord(inp, lst):
indx = lst.index(inp)
f_ans = lst[indx + 1]
return f_ans
def get_indx_purchase(inp, lst):
indx = int(lst.index(inp))
f_ans = ''
for i in lst[indx:indx+4]:
if i.isnumeric():
f_ans = i
break
else:
f_ans =''
return f_ans
def get_company_name(inp,lst):
idx = int(lst.index(inp))
f_ans = ''
f_indx = int(idx + 1)
for i in lst[f_indx]:
if i.isnumeric():
f_ans = lst[f_indx + 1]
else:
f_ans = lst[f_indx]
return f_ans
def get_company_name_2(sublist, lst):
f_ans = ''
print(sublist)
if "From:" in sublist:
f_ans = get_company_name("From:", lst)
print(f_ans)
elif "From" in sublist:
f_ans = get_company_name("From", lst)
elif "Invoice" in sublist:
f_ans = get_company_name("Invoice", lst)
else:
f_ans = lst[0]
return f_ans
def Get_shipping_cost(inp, lst):
f_ans = ''
inpt = ''
if inp:
inpt = inp[0]
else:
inpt = "0"
if inpt.isnumeric():
f_ans = inpt
else:
indx = lst.index(inpt)
f_indx = int(indx + 1)
f_ans = lst[f_indx]
if f_ans.isnumeric():
f_ans = lst[f_indx]
else:
f_ans = 0
return f_ans
order_2 = get_index_ord(nopA[0], flat_list)
company_name = get_company_name_2(qrsA, flat_list)
purchase_2 = get_indx_purchase("Invoice", flat_list)
def vat_2(inp):
ans_vat = ''
new_set = list(set([i[0] for i in inp]))
new_set.sort()
if new_set[-1] - new_set[-2] in new_set:
ans_vat = new_set[-1] - new_set[-2]
else:
ans_vat = 0
return ans_vat
vat_2ans = vat_2(defgA)
shipping = int(Get_shipping_cost(uvwA, flat_list))
# print(shipping)
d["date"] = min(abcA)
d["total_amount"] = max(defgA)[0]
d["vat"] = vat_2ans - shipping if int(hijA[0][0:-1]) /100 * d["total_amount"] == 0 else int(hijA[0][0:-1]) /100 * d["total_amount"] - shipping
d["invoice_number"] = klmA[0] if klmA[0] else purchase_2
d["purchase_order_number"] = nopA[0] if nopA[0].isnumeric() else order_2
d["supplier_name"] = company_name
# print("Apiyo", d)
return d
from azure.cognitiveservices.vision.computervision import ComputerVisionClient
from azure.cognitiveservices.vision.computervision.models import OperationStatusCodes
from azure.cognitiveservices.vision.computervision.models import VisualFeatureTypes
from msrest.authentication import CognitiveServicesCredentials
import time
subscription_key = "0f5bb668eab84237bd5742644f3cae02"
endpoint = "https://ocr-test-for-python.cognitiveservices.azure.com/"
computervision_client = ComputerVisionClient(
endpoint, CognitiveServicesCredentials(subscription_key))
remote_image_handw_text_url = "0.jpg"
# Call API with URL and raw response (allows you to get the operation location)
recognize_handw_results = computervision_client.read(
remote_image_handw_text_url, raw=True)
# Get the operation location (URL with an ID at the end) from the response
operation_location_remote = recognize_handw_results.headers[
"Operation-Location"]
# Grab the ID from the URL
operation_id = operation_location_remote.split("/")[-1]
# Call the "GET" API and wait for it to retrieve the results
while True:
get_handw_text_results = computervision_client.get_read_result(
operation_id)
if get_handw_text_results.status not in ['notStarted', 'running']:
break
time.sleep(1)
mlpreview(url)
# This requires two calls using batch_read_file() and
# get_read_operation_result(). The call to batch_read_file() is
# asynchronous. In the results of the call to
# get_read_operation_result(), we need to check if the first call
# completed with OperationStatusCodes before extracting the text
# data. The results include the text as well as the bounding box
# coordinates for the text.
raw = True
numberOfCharsInOperationId = 36
# Asynchronous call.
try:
rawHttpResponse = client.read(url, raw=raw)
except Exception as e:
catch_exception(e, url)
# Get ID from returned headers.
operationLocation = rawHttpResponse.headers["Operation-Location"]
idLocation = len(operationLocation) - numberOfCharsInOperationId
operationId = operationLocation[idLocation:]
# Wait for the result.
while True:
result = client.get_read_result(operationId)
if result.status not in [
OperationStatusCodes.not_started, OperationStatusCodes.running
'''
END - Authenticate
'''
# <snippet_read_call>
'''
OCR: Read File using the Read API, extract text - remote
This example will extract text in an image, then print results, line by line.
This API call can also extract handwriting style text (not shown).
'''
print("===== Read File - remote =====")
# Get an image with text
read_image_url = "https://raw.githubusercontent.com/MicrosoftDocs/azure-docs/master/articles/cognitive-services/Computer-vision/Images/readsample.jpg"
# Call API with URL and raw response (allows you to get the operation location)
read_response = computervision_client.read(read_image_url, raw=True)
# </snippet_read_call>
# <snippet_read_response>
# Get the operation location (URL with an ID at the end) from the response
read_operation_location = read_response.headers["Operation-Location"]
# Grab the ID from the URL
operation_id = read_operation_location.split("/")[-1]
# Call the "GET" API and wait for it to retrieve the results
while True:
read_result = computervision_client.get_read_result(operation_id)
if read_result.status not in ['notStarted', 'running']:
break
time.sleep(1)
'''
END - Batch Read File - local
'''
# <snippet_read_call>
'''
Batch Read File, recognize printed text - remote
This example will extract printed text in an image, then print results, line by line.
This API call can also recognize handwriting (not shown).
'''
print("===== Batch Read File - remote =====")
# Get an image with printed text
remote_image_printed_text_url = "https://raw.githubusercontent.com/Azure-Samples/cognitive-services-sample-data-files/master/ComputerVision/Images/printed_text.jpg"
# Call API with URL and raw response (allows you to get the operation location)
recognize_printed_results = computervision_client.read(
remote_image_printed_text_url, language='en', raw=True)
# </snippet_read_call>
# <snippet_read_response>
# Get the operation location (URL with an ID at the end) from the response
operation_location_remote = recognize_printed_results.headers[
"Operation-Location"]
# Grab the ID from the URL
operation_id = operation_location_remote.split("/")[-1]
# Call the "GET" API and wait for it to retrieve the results
while True:
get_printed_text_results = computervision_client.get_read_result(
operation_id)
if get_printed_text_results.status not in ['NotStarted', 'Running']:
break
# Set credentials
credentials = CognitiveServicesCredentials(key)
# Create client
client = ComputerVisionClient(endpoint, credentials)
# change this URL to reflect the image that you would like to test.
url = "https://azurecomcdn.azureedge.net/cvt-181c82bceabc9fab9ec6f3dca486738800e04b45a0b3c1268609c94f4d67173a/images/shared/cognitive-services-demos/analyze-image/analyze-6-thumbnail.jpg"
# image_path = "images/computer_vision_ocr.png"
lang = 'en'
raw = True
custom_headers = None
# Read an image from a url
rawHttpResponse = client.read(url, language=lang, custom_headers=custom_headers, raw=raw)
# Uncomment the following code and comment out line 37 to read from image stream
# with open(image_path, "rb") as image_stream:
# rawHttpResponse = client.read_in_stream(
# image=image_stream, language=lang,
# # Raw will return the raw response which can be used to find the operation_id
# raw=True
# )
# Get ID from returned headers
operationLocation = rawHttpResponse.headers["Operation-Location"]
operationId = operationLocation.split('/')[-1]
# SDK call
while True: