将Google Vision API响应转换为JSON

Kus*_*ikh 5 python json google-api

任务:

  • 将Google Vision API响应转换为JSON

问题:

  • API调用的返回值不是JSON格式

Python函数

def detect_logos(path):
"""Detects logos in the file."""
client = vision.ImageAnnotatorClient()

# [START migration_logo_detection]
with io.open(path, 'rb') as image_file:
    content = image_file.read()

image = types.Image(content=content)

response = client.logo_detection(image=image)
logos = response.logo_annotations

print('Logos:')
print(logos)
print(type(logos))
Run Code Online (Sandbox Code Playgroud)

Google在线JSON

"logoAnnotations": [
{
  "mid": "/m/02wwnh",
  "description": "Maxwell House",
  "score": 0.41142157,
  "boundingPoly": {
    "vertices": [
      {
        "x": 74,
        "y": 129
      },
      {
        "x": 161,
        "y": 129
      },
      {
        "x": 161,
        "y": 180
      },
      {
        "x": 74,
        "y": 180
      }
    ]
  }
}
Run Code Online (Sandbox Code Playgroud)

Google回应(清单)

 [mid: "/m/02wwnh"
description: "Maxwell House"
score: 0.4114215672016144
bounding_poly {
  vertices {
    x: 74
    y: 129
  }
  vertices {
    x: 161
    y: 129
  }
  vertices {
    x: 161
    y: 180
  }
  vertices {
    x: 74
    y: 180
  }
}
]
Run Code Online (Sandbox Code Playgroud)

类型:

google.protobuf.internal.containers.RepeatedCompositeFieldContainer

尝试过:

在Python中将Protobuf转换为JSON

小智 7

Google Vision 2.0 需要不同的代码,如果不更改代码,将抛出以下错误:

object has no attribute 'DESCRIPTOR'
Run Code Online (Sandbox Code Playgroud)

以下是如何使用 json 和/或 protobuf 序列化和反序列化的示例:

import io, json
from google.cloud import vision_v1
from google.cloud.vision_v1 import AnnotateImageResponse

with io.open('000048.jpg', 'rb') as image_file:
    content = image_file.read()

image = vision_v1.Image(content=content)
client = vision_v1.ImageAnnotatorClient()
response = client.document_text_detection(image=image)

# serialize / deserialize proto (binary)
serialized_proto_plus = AnnotateImageResponse.serialize(response)
response = AnnotateImageResponse.deserialize(serialized_proto_plus)
print(response.full_text_annotation.text)

# serialize / deserialize json
response_json = AnnotateImageResponse.to_json(response)
response = json.loads(response_json)
print(response['fullTextAnnotation']['text'])
Run Code Online (Sandbox Code Playgroud)

注1:proto-plus 不支持转换为snake_case 名称,protobuf 支持“preserving_proto_field_name=True”。因此,目前无法将字段名称从 response['full_text_annotation'] 转换为 response['fullTextAnnotation'] 对此有一个开放的功能请求:googleapis/proto-plus-python#109

注 2:如果 x=0,google vision api 不会返回 x 坐标。如果 x 不存在,protobuf 将默认 x=0。在使用 MessageToJson() 的 python vision 1.0.0 中,这些 x 值未包含在 json 中,但现在在 python vision 2.0.0 和 .To_Json() 中,这些值包含为 x:0


小智 6

该库返回普通的 protobuf 对象,可以使用以下方法将其序列化为 JSON:

from google.protobuf.json_format import MessageToJson
serialized = MessageToJson(original)
Run Code Online (Sandbox Code Playgroud)

这对我有用。

  • 它返回此错误:`AttributeError: 'google.protobuf.pyext._message.RepeatedCompositeCo' object has no attribute 'DESCRIPTOR'` (12认同)
  • `MessageToJson(response._pb)` 对我有用。[归功于 Martin J](/sf/ask/4508261621/) (4认同)

Kus*_*ikh 5

找到解决方案。它不能转换为 JSON,但可以像这样访问:

print(logos[0].bounding_poly.vertices[0].x)
Run Code Online (Sandbox Code Playgroud)