ABSTRACT
This chapter provides the different parts of a vision system and the way to manage them optimally: some basic image analysis techniques such as grabbing of a picture, thresholding, slicing, windowing, and filtering operations: and the possible applications of vision techniques to fish or fish fillet inspection and the problem met in these applications. Improving the histogram quality through the introduction of filters or through an adapted background choice makes it possible to avoid time-consuming software operations during the isolation of a defect. The fish fillet is entirely in the image, and the background occupies the whole remaining room. The result of a thresholding operation is to differentiate pixels with higher and lower values than a threshold in an image. Enhance the contrast between the fillet, the defect and the background, through the use of filters; the aim is to obtain more distinguishable peaks in the histogram.
