Open AccessAutomatic Segmentation of Skin Regions in Thermographic Images: an Experimental Study
Author(s) -
Cristina M. R. Roseiro,
Luís Roseiro
Publication year - 2021
Publication title -
wseas transactions on signal processing
Language(s) - English
Resource type - Journals
eISSN - 2224-3488
pISSN - 1790-5052
DOI - 10.37394/232014.2021.17.7
Subject(s) - thermography , closing (real estate) , skin temperature , segmentation , computer science , artificial intelligence , computer vision , skin lesion , identification (biology) , action (physics) , biomedical engineering , pattern recognition (psychology) , infrared , medicine , pathology , physics , law , political science , optics , botany , quantum mechanics , biology
Infrared thermography can be applied in medical applications, such as monitoring skin temperature in inflammatory processes. The possibility for health care professionals and patients to be able to easily, quickly and economically, at anytime and anywhere, monitor the skin temperature distribution through the acquisition of images to control skin infections is extremely important nowadays. This work aims to develop an automatic methodology for the segmentation, identification, analysis and diagnosis of skin inflammation based on thermographic images. The study compares thermographic images from subregions of the hand skin and presents an experimental investigation to segment and identify features in the images automatically. Left and righthand images from two volunteers’ obtained in different conditions, such as cold action, activity action (opening and closing the hand), and friction action (rub both hands), were considered and analyzed. The obtained results demonstrate the feasibility of the implemented procedures and encourage developing and implementing an operating system to monitor skin infections in thermographic images.