At the end of 2016 year Microsoft announced the availability of a set of 21 new APIs called Microsoft Cognitive Services. Cognitive Services provides simple APIs that handle common use cases, such as recognizing speech or performing facial recognition on an image. These APIs can be broken down into five main categories: vision, speech, language, knowledge, and search. They are all based on machine learning algorithms.
In this blog post, I’m going to explore the Computer Vision API, which returns rich information about visual content found in an image.
The Computer Vision API has many features that include the ability to analyze a picture to understand its content, create smart thumbnails (to ensure you never crop the region of interest), as well as OCR and adult content detection.
Adult content detection is one feature of the Vision APIs, for example by one API call we can have a probability score for inappropriate content within the picture and decide the action to take, for example we can decide to not stored in the cloud the image with adult content.
To use the Computer Vision API you have registering for a Cognitive Service API on Microsoft Azure platform, you can start trying it by a free account:
Cognitive Services Vision APIs, provide a list of features you wish to enable in the analysis process, the options available to you are:
I tested the Computer Vision API in a scenario regarding my children: to check if their room are okay or not 🙂
Do you hate discussions when it comes to the topic of cleaning? Because children perceive mess differently, they feel treated unfairly and are frustrated themselves.
Wouldn’t it be nice to have a little tool that shows children, whether their rooms are okay or not? And that shows them, at which places they need to clean better? That is why I used the Microsoft Computer Vision API for room kids check !
Thanks to latest image processing technology, I used to Computer Vision API to measure the tidiness in the room and displays the result on scale from 0 (perfectly tidy) to 10 (chaos).
My webcam placed in the room acquire the images from the kids room, for example in this photo there was a perfectly tidy room:
meanwhile in this photo there was a little of mess:
I used a Node-RED flow to acquire image from camera, invoke the Computer Vision rest API and to assign a score to the tidiness in the room and displays the result on scale from 0 (perfectly tidy) to 10 (chaos). My algorithm assign a score based on the different items contained in description.
Here are some examples:
Fortunately this is not my kids room, the description of this image says it all, really:
Microsoft Cognitive Services allows developers to build incredibly complex features into apps using only a couple of lines of code.
I have a great time with these APIs.