Test Driving Google Cloud Vision With a Raspberry Pi Robot

In our most recent binge of projects at Dexter Industries, we have begun to use the Google Cloud Vision API on the Raspberry Pi. Google Cloud Vision is a powerful API designed to allow you to process, generate context, and analyze images using Google's Machine Learning technology.

Machine learning powered image processing has wide ranging applications in robotics, and promises a revolution in many industries over the next decade, from agriculture, to transportation, to manufacturing: vision guided robots will play an increasingly important part of new technologies in this areas.

We wanted to be able to develop robots (and products) that use Google's Cloud Vision API to process images using the Raspberry Pi, the compact $35 credit-card-sized-computer that's become widely popular for use in education, engineering, and technology. In the first in a series of how to use Google's Cloud Vision API with the Raspberry Pi we walk you through how to setup the service on your Pi, and how to use three of the most powerful services with the Raspberry Pi Camera:

• Image Labeling - Labeling an image for contents and context.
• Facial Analysis - Analyzing the location, position, and emotional traits of faces.
• Logo Detection - Detecting organization logos in pictures.

In our first project, we use Google Cloud Vision to analyze and  Sort Halloween Candy. While the task may seem simple, that’s exactly the point. This tutorial demonstrates a really simple but boring task that can be automated with Google Cloud Vision. Making this technology accessible is important, and combining Google Cloud Vision and the Raspberry Pi Camera does that!

The robots are coming, and they’ll probably be running Google Cloud Vision!

You can see the full tutorial on how to use Google Cloud Vision On the Raspberry Pi here.