如何为 OCR 制作 REST API

How to make a REST API for OCR

我正在执行一项任务,其中我有一张图像并且必须从中提取日期。使用 Google Cloud Platform Vision API 提取日期。如何使用接受图像和 returns 日期的烧瓶制作 API?格式如下所示:

Request:POST /extract_date

Payload: {“base_64_image_content”: }

Response: If date is present: {“date”: “YYYY-MM-DD”} If date is not present: {“date”: null}

你能帮帮我吗?

Flask is among the most popular web frameworks for Python. It's relatively easy to learn, while its extension Flask-RESTful 使您能够快速构建 REST API。

最小示例:

from flask import Flask
from flask_restful import Resource, Api

app = Flask(__name__)
api = Api(app)

class MyApi(Resource):
    def get(self, date):
        return {'date': 'if present'}

api.add_resource(MyApi, '/')

if __name__ == '__main__':
    app.run()

测试 curl:

curl http://localhost:5000/ -d "data=base_64_image_content" -X PUT

经过评论中的讨论,以下是如何使用 GCP Functions 构建 OCR REST API:

import re
import json
from google.protobuf.json_format import MessageToJson
from google.cloud import vision
from flask import Response


def detect_text(request):
    """Responds to any HTTP request.
    Args:
        request (flask.Request): HTTP request object.
    Returns:
        The response text or any set of values that can be turned into a
        Response object using
        `make_response <http://flask.pocoo.org/docs/0.12/api/#flask.Flask.make_response>`.
    """
    client = vision.ImageAnnotatorClient()
    image = vision.types.Image(content=request.data)
    response = client.text_detection(image=image)
    serialized = MessageToJson(response)
    annotations = json.loads(serialized)

    full_text = annotations['textAnnotations'][0]['description']
    annotations = json.dumps(annotations)

    r = Response(response=annotations, status=200, mimetype="application/json")
    return r

这是您可以用来发出请求的一段代码:

def post_image(path, URL):
    headers = {'content-type': 'image/jpeg'}
    img = open(path, 'rb').read()
    response = requests.post(URL, data=img, headers=headers)
    return response