Image analysis based on fuzzy logic

  • Authors:
  • Catalin Gheorghe Amza

  • Affiliations:
  • Materials Technology and Welding Department, University Politehnica of Bucharest, Bucharest, Romania

  • Venue:
  • VIS'08 Proceedings of the 1st WSEAS international conference on Visualization, imaging and simulation
  • Year:
  • 2008

Quantified Score

Hi-index 0.00

Visualization

Abstract

This paper presents the use of artificial intelligence techniques such as fuzzy logic for non-destructive X-ray testing of food products. The proposed original algorithm detects the presence of defects in a food product in two stages. A radiographic image obtained from the inspected product is first segmented into meaningful objects and then analyzed by a two stage fuzzy logic algorithm. The first stage of the algorithm checks whether a possible defect (an area from the radiographic image) corresponds from the geometrical point of view (shape, size, etc., characteristics that depends entirely of the nature of the product that is inspected). If the possible defect corresponds form the geometrical point of view, then the second stage of detection is applied. This stage verifies whether the possible defect fulfills some "logical" criteria. These criteria are based on grey-level statistics for the corresponding area in question (i.e. between the possible defect area and the surrounding background there is a contrast difference of a certain level).