Neural network edge detectors for separation of particles in 2-D grey-scale images

A method is developed for edge detection in grey-scale images of particles using artificial neural network classifiers. The edge detection is used to separate the individual particles in the image especially where the particles are touching each other. Once the particles are separated individual measurements can be made to compile accurate size-distribution information. Images of eighty calibrated gravel stones are used to test the method. Gravel stones are adopted as abstracts for the general class of irregularly shaped particles. Comparisons are made between probabilistic neural network multilayer perceptron and Gaussian mixture model Bayesian classifiers for edge detection. Methods are developed to reduce the number of feature vectors and features in order to speed up the edge defection process without a loss in performance for the probabilistic neural network edge detector