(Image Credit: iStockPhoto/Dimitrios Stefanidis)

When it comes to machine learning, few companies can match the investment and experience of Google. Their powerful technology is used across the company’s own beloved products and services to improve the experience of their users – and now they’re opening up more of it for any developer to use in their own applications. 

Announced today, Google is releasing a public beta of its Cloud Vision API which provides image recognition and classification smarts to applications. Its abilities range from the simple – like OCR text recognition – to recognising exact objects within an image like we’ve seen used in Google Photos. 

In Google’s photos app – users can find their photos automatically organised into separate categories based on what Google’s smart algorithms have detected such as; people, animals, flowers, and landmarks. This categorisation is able to go a lot deeper, and can even be used to keep things in your app PG-rated. 

When curating user-generated photos, it helps to remove some of the unpredictability and enable more control over the content you display. For example, you may only want to show photos of happy people within your app to help invoke that same feeling in your users – which can be achieved through ‘sentiment analysis’ in the API. 

(Image: Sentiment analysis)

Cloud Vision API supports a broad set of scenarios from: 

• Insights from your images: Powered by the same technologies behind Google Photos, Cloud Vision API detects broad sets of objects in your images — from flowers to popular landmarks 
• Inappropriate content detection: Powered by Google SafeSearch, Cloud Vision API moderates content from your crowd sourced images by detecting different types of inappropriate content. 
• Image sentiment analysis: Cloud Vision API can analyze emotional attributes of people in your images, like joy, sorrow and anger, along with detecting popular product logos. 
• Text extraction: Optical Character Recognition (OCR) enables you to detect text within your images, along with automatic language identification across a broad set of languages. 

This functionality isn’t free, however, and Google has announced various tiers of pricing dependent on what you need the API to do. Basic features, like OCR, will cost just $0.60 per 1,000 images – whereas more technical features such as label detection will cost $2 per 1,000 images. None of the options are particularly expensive considering what you’re getting for your bucks, although it’s more pricey than Microsoft’s similar Project Oxford. 

It’s worth noting that, during the beta, Google is restricting users to 20 million images per month. This will be more than enough for most developers, but it’s a limitation nonetheless. Google notes that some companies are using the API for their products already; including Yik Yak who has implemented Cloud Vision for text extraction and feature detection.

Find more information and get started with the Cloud Vision API here. 

Have you tried Google’s Cloud Vision API yet? Share your experience in the comments.