Open CV is an artificial intelligence library related to computer vision. This post talks about how we used OpenCV to assist analysis of certain medical images.
OpenCV
OpenCV is a library written in C++ with bindings for Python and Java which makes it easier to apply to digest and process images.
OpenCV is designed to do things like:
- 2D and 3D feature toolkits
- Facial recognition system
- Gesture recognition and motion tracking
- Mobile robotics
- Motion understanding
- Object detection
OpenCV also provides and uses a variety of interesting statistical algorithms that people associate with AI including things like k-nearest neighbor algorithm, Naive Bayes classifier, artificial neural networks, random forest, support vector machine, deep neural networks, etc.
Applying OpenCV
What we actually did with OpenCV was quite simple actually. We were processing images from a medical device that already highlighted certain portions in a blue color. What we needed to do was to analyze how much of the image was blue over a number of samples.
So our code just used python to iterate over the image files and then used opencv on each file to essentially determine what % was within a range of the blue we were looking for.
The output was images classified based on the amount of blue, which was thought to correlate with other hypotheses.
In theory, if it works the way we think, the detection of trends in the % of blue could help to classify or diagnose certain conditions.
Cool, right!?
References
Matt is a principal.