Textextraktion (Lese-API) mit Azure Computer Vision-API (Python3.6)

Einführung

Ich habe versucht, Text aus dem Bild zu extrahieren

Entwicklungsumgebung

• Windows 10
• Anaconda
• Python 3.6
• OpenCV 4.4.0
• Azure Computer Vision API --Computer Vision Client Library (falls erforderlich)

Einführung

1. 1. Melden Sie sich beim Azure-Portal an (https://portal.azure.com/).

2. Erstellen Sie eine Ressource für die Computer Vision-API

3. 3. Notieren Sie sich den Schlüssel und den Endpunkt

Vier. Bitte installieren Sie die erforderlichen Bibliotheken.

pip install matplotlib
pip install pillow
pip install opencv-python
pip install --upgrade azure-cognitiveservices-vision-computervision 

Fünf. Geben Sie den Schlüssel und den Endpunkt ein, den Sie notiert haben, und führen Sie den folgenden Code aus!

subscription_key = "<your subscription key>"
endpoint = "<your API endpoint>"

Der Endpunkt scheint auch dann zu funktionieren, wenn Sie die Region angeben.

endpoint = "https://<your region>.api.cognitive.microsoft.com/"

Extrahieren Sie Text aus der Bild-URL

[Schnellstart: Extrahieren Sie gedruckten und handgeschriebenen Text mit der REST-API und Python von Computer Vision](https://docs.microsoft.com/en-us/azure/cognitive-services/computer-vision / quickstarts / python-hand-text)

import json
import os
import os.path
import sys
import requests
import time
import matplotlib.pyplot as plt
from matplotlib.patches import Polygon
from PIL import Image
from io import BytesIO
 import cv2

subscription_key = "<your subscription key>"
endpoint = "<your API endpoint>"
 endpoint = "https://japanwest.api.cognitive.microsoft.com/"
text_recognition_url = endpoint + "vision/v3.1/read/analyze"

image_url = "https://raw.githubusercontent.com/MicrosoftDocs/azure-docs/master/articles/cognitive-services/Computer-vision/Images/readsample.jpg "
headers = {'Ocp-Apim-Subscription-Key': subscription_key}
data = {'url': image_url}
response = requests.post(text_recognition_url, headers=headers, json=data)
response.raise_for_status()

operation_url = response.headers["Operation-Location"]
analysis = {}
poll = True
while (poll):
    response_final = requests.get(response.headers["Operation-Location"], headers=headers)
    analysis = response_final.json()
    
    print(json.dumps(analysis, indent=4))

    time.sleep(1)
    if ("analyzeResult" in analysis):
        poll = False
    if ("status" in analysis and analysis['status'] == 'failed'):
        poll = False

polygons = []
if ("analyzeResult" in analysis):
    polygons = [(line["boundingBox"], line["text"])
                for line in analysis["analyzeResult"]["readResults"][0]["lines"]]

image = Image.open(BytesIO(requests.get(image_url).content))
ax = plt.imshow(image)
for polygon in polygons:
    vertices = [(polygon[0][i], polygon[0][i+1])
                for i in range(0, len(polygon[0]), 2)]
    text = polygon[1]
    patch = Polygon(vertices, closed=True, fill=False, linewidth=2, color='y')
    ax.axes.add_patch(patch)
    plt.text(vertices[0][0], vertices[0][1], text, fontsize=20, va="top")
plt.show()

input	output

Extrahieren Sie Text aus dem lokalen Bild

import json
import os
import os.path
import sys
import requests
import time
import matplotlib.pyplot as plt
from matplotlib.patches import Polygon
from PIL import Image
from io import BytesIO
import cv2

subscription_key = "<your subscription key>"
endpoint = "<your API endpoint>"
 endpoint = "https://japanwest.api.cognitive.microsoft.com/"
text_recognition_url = endpoint + "vision/v3.1/read/analyze"

headers = {'Ocp-Apim-Subscription-Key': subscription_key, 'Content-Type': 'application/octet-stream'}
filename = "readsample.jpg "
root, ext = os.path.splitext(filename)
 image_data = open(filename, "rb").read()
color = cv2.imread(filename, cv2.IMREAD_COLOR)
cv2.namedWindow("color", cv2.WINDOW_NORMAL)
cv2.imshow("color", color)
cv2.waitKey(1)
image_data = cv2.imencode(ext, color)[1].tostring()
response = requests.post(text_recognition_url, headers=headers, data=image_data)
response.raise_for_status()

operation_url = response.headers["Operation-Location"]
analysis = {}
poll = True
while (poll):
    response_final = requests.get(
        response.headers["Operation-Location"], headers=headers)
    analysis = response_final.json()
    
    print(json.dumps(analysis, indent=4))

    time.sleep(1)
    if ("analyzeResult" in analysis):
        poll = False
    if ("status" in analysis and analysis['status'] == 'failed'):
        poll = False

polygons = []
if ("analyzeResult" in analysis):
    polygons = [(line["boundingBox"], line["text"])
                for line in analysis["analyzeResult"]["readResults"][0]["lines"]]

 image = Image.open(BytesIO(image_data))
image = Image.fromarray(color)
ax = plt.imshow(image)
for polygon in polygons:
    vertices = [(polygon[0][i], polygon[0][i+1])
                for i in range(0, len(polygon[0]), 2)]
    text = polygon[1]
    patch = Polygon(vertices, closed=True, fill=False, linewidth=2, color='y')
    ax.axes.add_patch(patch)
    plt.text(vertices[0][0], vertices[0][1], text, fontsize=20, va="top")
plt.show()

input	output

Verwenden Sie die Computer Vision-Clientbibliothek

Schnellstart: Verwenden Sie die Computer Vision Client Library (https://docs.microsoft.com/en-us/azure/cognitive-services/Computer-vision/quickstarts-sdk/client-library?pivots=programming-language) -python & tabs = Visual-Studio)

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

from array import array
import os
from PIL import Image
import sys
import time
import cv2 
from io import BytesIO

subscription_key = "<your subscription key>"
endpoint = "<your API endpoint>"
 endpoint = "https://japanwest.api.cognitive.microsoft.com/"

computervision_client = ComputerVisionClient(endpoint, CognitiveServicesCredentials(subscription_key))

print("===== Batch Read File - remote =====")
remote_image_handw_text_url = "https://raw.githubusercontent.com/MicrosoftDocs/azure-docs/master/articles/cognitive-services/Computer-vision/Images/readsample.jpg "

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)

if get_handw_text_results.status == OperationStatusCodes.succeeded:
    for text_result in get_handw_text_results.analyze_result.read_results:
        for line in text_result.lines:
            print(line.text)
            print(line.bounding_box)
print()

===== Batch Read File - remote =====
The quick brown fox jumps
[38.0, 650.0, 2572.0, 699.0, 2570.0, 854.0, 37.0, 815.0]
over
[184.0, 1053.0, 508.0, 1044.0, 510.0, 1123.0, 184.0, 1128.0]
the lazy dog!
[639.0, 1011.0, 1976.0, 1026.0, 1974.0, 1158.0, 637.0, 1141.0]

Danke für deine harte Arbeit.