Ulcers are chronic wounds that cause severe pain and discomfort to the patients. Describing the ulcer tissues in terms of percentages of each tissue colour is an approved method of wound healing assessment. The first indication of ulcer healing is the growth of the red granulation tissue from the base of ulcers. Granulation tissue appears red in colour due the hemoglobin content in the blood capillaries. The main objective of this research work is to investigate the optical characteristics of hemoglobin content in ulcers as a possible image marker in detecting granulation tissue from RGB colour images of chronic ulcers. Independent Component Analysis is implemented to extract grey-level hemoglobin images from RGB colour images of chronic ulcers. Extracted hemoglobin images reflect areas of hemoglobin distribution representing detected regions of granulation tissue. K-means clustering is implemented to classify and segment detected regions of granulation tissue. Preliminary analysis of the results indicates an overall accuracy of 97.31 of the algorithm performance when compared to manual segmentation. The ultimate aim of this research work is to produce an objective non-invasive approach that aid physicians to assess the healing status of chronic ulcers in a more precise and reliable way. © 2011 IEEE.