From Human Vision to Image Processing
DThong Nguyen · Transactions of the Institution of Professional Engineers New Zealand: Electrical/Mechanical/Chemical Engineering Section · 1988
The goal of image processing, whether it is for computer vision applications or for image-coding for visual communications, is to extract characteristic features and to discard redundancy in the image. If the processing scheme can imitate the human visual system, then high efficiency can be achieved. This semi-tutorial paper presents recent developments and theories on the human visual mechanism, then relates these to progress in image coding and computer vision techniques. The two main interpretations of the mechanism in the early visual cortex: the feature-detector and the Fourier analyser, will be discussed in the context of edge detection. Some preliminary results of our research work reveal that the best edge detection technique is a compromise between the two interpretations.