Google Cloud Vision Serverless Implementation

Google Cloud Vision Hello World Image

Google Cloud Vision API is an image processing Google cloud service to detected and label images metadata and information. It can detect Faces, Logos, Landmark, and Text from images with predefined machine learning models with great accuracy.

In this article, We are using Google Cloud Functions to provide an endpoint to upload images and get the response.

Google Cloud Functions Implementation

git clone https://github.com/vinaymavi/Google-Cloud-Vision-Serverless-Implementation

Google Cloud Vision Serverless Implementation

Pre-Requisite — Google cloud sdk setup — https://cloud.google.com/sdk/install

Deploy to Google Cloud Functions:
$ gcloud functions deploy vision_serverless — runtime nodejs8 — trigger-http — project <your_project_name>
Curl Request
curl -X POST \
<cloud_function_url> \
 -H ‘Content-Type: application/json’ \
 -H ‘Postman-Token: e41268f6–9660–4be9-b5da-a99ad78689f9’ \
 -H ‘cache-control: no-cache’ \
 -d ‘{
 “base64Image”:”<image_base64>”
}’

Response

Google Cloud Vision Serverless Implementation output.

Reference:

• https://github.com/vinaymavi/Google-Cloud-Vision-Serverless-Implementation
• https://cloud.google.com/vision/
• https://cloud.google.com/nodejs/docs/reference/vision/0.24.x/
• https://cloud.google.com/functions/