Camouflage Defect Identification: A Novel Approach
Authors:
- 1 December 2006
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 145-148
Abstract
Camouflaging is an attempt to obscure the signature of a target and sublimate the target in its background frame in terms of co-occurrence matrix texture properties. The objective of this research work is to detect such camouflaged defect. Since the texture similarity between the background and the defect is the premise for accomplishing camouflage, an intensive texture analysis should be able to discriminate defect from its background. In the proposed work we intended to employ Co-occurrence matrix based texture features are computed within the block, a small region of the image. Finally defective portions are detected by cluster analysis and identified through!. Watershed segmentation. Several experiments have been conducted to demonstrate the suitability of the proposed model to spot camouflaged defects. Experimentation on the synthetic and real images corroborates the potential of the method.Keywords
This publication has 1 reference indexed in Scilit:
- Surface texture analysis based upon the visually acquired perturbation of surface normalsImage and Vision Computing, 1997