Optical Hough And Fourier Processors For Product Inspection

Abstract

Coherent optical processors for product inspection are discussed, and data from a case study are presented. An optical processor that produces both Fourier and Hough transform data is presented. The Fourier transform data are wedge /ring sampled to produce in -plane distortion-invariant features to determine the product's attribute (good or defective) and its orientation. Hough transform data are used for new quantitative mensuration information on good products. In one case, optical Hough transforms are quantitatively shown to yield more consistent results than do digital realizations.

<title>Neural network classification of Fraunhofer diffraction patterns for inspection of fine-pitch electronic components</title>

“…Preprocessing the moments into the following quantities allows the data to be contained to a range suitable for input to the classifier: a=•- (5) Noting that the symmetry or lack thereof(skewness) ofthe halfdiffraction pattern is measured by means of the quantity: a3=- (6) Also, the extent to which the pauern is peaked or flat is measured by means ofthe kurtosis parameter: a4=- (7) Each diffraction pattern can be represented by a feature vector consisting of twelve parameters, c to a13. number and nature of parameters have not arbitrarily been chosen, they are as a result of experimentation.…”

A Survey of Techniques in Aided Target Recognition (ATR) from Digital and Optical Perspectives

“…Richards [14] compare recognition results of using both the HT and FT on identical image data in a product inspection application. Two optical HT architectures were examined, differing in the choice of transform variables (one Cartesian, the other poiar coordinates).…”

Spatial Light Modulator Used for the Fourier Transformation of Video Images

“…An overview was given by Efron [6]. With a combination of a video camera and liquid crystal television (LCTV) device Casasent and colleagues [7][8][9] constructed a commercially available real-time (15 objects s -1) Fourier transform inspection system.…”