An evaluation study of traditional and neural network techniques for image processing applications

A comparative study of the artificial neural computing and traditional approaches to image processing is performed. The major goal is to determine the usefulness of artificial neural systems (ANSs) for such image processing applications as histogramming and image encoding. The paradigm used was developed from a C programming language model of a perceptron ANS with consideration of backpropagation attributes. It is found that the ANS approach produces results similar to those of traditional techniques.<
>