A rule-based, adaptive window-size filter for the enhancement of subcutaneous vascular patterns in thermographic images

A novel adaptive filtering algorithm has been developed. By adapting the filtering protocol and window size to local pixel populations, the rule-based filtering algorithm minimizes edge-broadening effects and enhances vascular boundaries. From the adaptively filtered result, thermographic subcutaneous vascular patterns can be extracted using a digital subtraction technique. This algorithm enhances faint vascular signals, as is evidenced by the clear delineation of the branching vein on the right forearm.<>