GENERATING IMAGE FILTERS FOR TARGET RECOGNITION BY GENETIC LEARNING

We describe results obtained from applying genetic algorithms to the problem of detecting targets in image data. The method we describe is a two-layered approach, with the first layer providing a focus-of-attention function for the second layer. The first layer is called a Screener and selects subimages from the original image data to be processed by the second layer, called the Classifier. The Screener reduces the computational load of the system. Each layer consists of a set of linear operators (filters) applied directly to the image data. A genetic algorithm is applied to populations of filters based on fitness criteria. We note that the statistical classifier chosen for the Classifier stage drives the evolution of filters that are useful for that classifier to make good discriminations.