pil example BookReader (snippet)

Python pil example 'BookReader'

Functions in program:

def get_text_from_return(res_json):
def google_cloud_vision (image_content):

Modules used in program:

import speech
import photos
import console
import Image
import io
import base64
import json
import requests

python BookReader

Python pil example: BookReader

import requests
import json
import base64
import io
import Image
import console
import photos
import speech

GOOGLE_CLOUD_VISION_API_URL = 'https://vision.googleapis.com/v1/images:annotate?key='
API_KEY = "AIzaSyDXY3DpMvWawvOd133LJR0iqOXD7Vx1T5Y"

def google_cloud_vision (image_content):
    api_url = GOOGLE_CLOUD_VISION_API_URL + API_KEY
    req_body = json.dumps({
        "requests": [{
            "image":{
                "content": image_content
            },
            "features":[{
                "type":"TEXT_DETECTION",
                "maxResults":10
            }]
        }]
    })
    res = requests.post(api_url, data=req_body)
    return  res.json()

#jsonからtext部分を取り出す
def get_text_from_return(res_json):
    text = res_json["responses"][0]["fullTextAnnotation"]["text"]
    return text

console.alert("ページを写して","","OK!")
pil_img = photos.capture_image()

size = (int(pil_img.size[0]*0.2), int(pil_img.size[1]*0.2))
pil_img = pil_img.resize(size)

#空のインスタンスに保存し、バイナリデータを取得する
img = io.BytesIO()
pil_img.save(img, "JPEG")
img_byte = img.getvalue()

img_str = str(base64.b64encode(img_byte), "utf-8")

#cloud vision apiで文字を認識させてtextデータを取得する
res_json = google_cloud_vision(img_str)
text_json = get_text_from_return(res_json)

##内容の確認と読み上げ
print(text_json)
speech.say(text_json)

Python links

Learn Python: https://pythonbasics.org/
Python Tutorial: https://pythonprogramminglanguage.com