Web Camera with Raspberry Pi3 to detect objects using Google Cloud Vision

Question

I want to integrate USB Web Camera with Raspberry Pi3 and send the images captured to Google Cloud Vision to detect objects. Any Python 3 library for doing the same?

I have successfully integrated my web camera and able to stream video over URL using "Motion"

Any library similar to Pi Camera or that can make me move forward from the above mentioned Motion library. would be of great help.

Answer 1

Done :)

import pygame
import pygame.camera
import time
import base64

from googleapiclient import discovery
from oauth2client.client import GoogleCredentials


pygame.camera.init()
pygame.camera.list_cameras()
cam = pygame.camera.Camera("/dev/video0", (640, 480))
cam.start()
time.sleep(0.1)
img = cam.get_image()
if file:
    pygame.image.save(img, file)
else:
    pygame.image.save(img, "img_captured.jpg")
cam.stop()

credentials = GoogleCredentials.get_application_default()
service = discovery.build('vision', 'v1', credentials=credentials)

with open('img_captured.jpg', 'rb') as image:
    image_content = base64.b64encode(image.read())
    service_request = service.images().annotate(body={
        'requests': [{
            'image': {
                'content': image_content.decode('UTF-8')
            },
            'features': [{
                'type': 'LOGO_DETECTION',
                'maxResults': 1
            }]
        }]
    })
    response = service_request.execute()

    try:
        label = response['responses'][0]['logoAnnotations'][0]['description']
    except:
        label = "No response."

    print("Detected -->" + label)