Illustrating physical principles through comparative feature extraction techniques in optical and digital image processing

摘要

Facilitation of pattern recognition in machine-vision systems or thematic mapper imagery can be used to illustrate physical principles through comparative feature extraction techniques in optical and digital image processing. This is demonstrated utilizing coherent image processing with Fourier transforms incorporating filtering techniques such as convolution and inverse Fourier transform. A parallel digital image processing sequence is applied to the same image. Transparencies of the final product image are optically processed with cross-correlation for comparison of results where applicable.