Knowledge-based system for on-line grading of herring roe

Grading of herring roe is a tedious and labor-intensive task. This paper presents techniques developed by us for automating the grading process. Visual and ultrasonic sensors are used to obtain information related to the size, shape, color and texture of each skein of roe. Computer vision techniques are developed to analyse the images and generate parametric features. Fuzzy reasoning that mimics human-expert reasoning, has been used to make decisions through a fuzzy logic rule base, which provides for accurate and consistent grading of bulk herring roe