Performance Measure and Efficiency of Thermal...

Performance Measure and Efficiency of Thermal Skin Burn Classification Using Image Processing Technique

Author Name : Yash G Pujari, Dr Malini Suvarna

Abstract

The objective of this research is to develop an automated method of determining the depth of Thermal Burn Wounds. Towards achieving this aim a database of Thermal Burn images has been created by collecting Burn images from hospitals, Doctors and the Internet.

The preprocessing involves context enhancement in lab color space by taking luminance component. Various pattern classifier techniques template matching (TMM) and artificial neural network has been applied on Thermal Burn images and performance comparison of the two techniques has been made. During the course of research. It has been observed that the ANN technique reflected the best result compare to TMM. The efficiency of the analysis and classification of the ANN technique has been of the order of 95% for grade -1 Burns, 97.5% for grade – 2 Burns and 95% for grade -3 Burns. As compared to 55% , 82.5% and 75% for TMM method. It is therefore felt that the ANN technique could be applied to analyzed and classify the Thermal Burns.

The Thermal Burn analysis technique could be safely used in remote location where specialist services are not readily available. This would definitely go a long way in mitigating the burn and suffering of the patients.

Key words: TMM, ANN