Towards a comprehensive assessment of wound-composition using color-image processing

This paper reports the results of our work on assessing wound-composition based on color-image processing. A knowledge-base (reference-clusters in the 3D color space) of different categories of wound-tissue and pigmentation was built, using about 9000 pixels from 48 images of various wounds. The HSI model was used for the purpose. We address 8 categories of wound-tissue and pigmentation, many more than those reported so far in the literature. The knowledge base was used subsequently for assessing wound-composition, through classification based on Mahalanobis distance. The results of experiments to test the efficacy of the algorithm are encouraging.