Feature extraction CNN algorithms for artificial immune systems

We introduce some CNN and analogic feature extraction algorithms for artificial immune systems, which are able to convert grayscale or color to binary images storing as much information as possible for further processing. We define a statistical property called immune histogram based on sub-patterns of these images. Our results and measurements show that these algorithms can be implemented in real-time applications. A sample application, which detects new textures in a familiar environment, is also presented.