Extended Machine Learning Model for the Burn Detection

Abstract

Burn identification is very important task for the sevieriety of the burn. The papre aims with the objective that the proper identification of the burn and making an automated framework for the identification of burn that support this real life problem. An extended model is designed in sucha manner that can classifiy the burn images into three class, which is done by mixed approach of digital image processing and machine learning. The proposed model performs better with a state-of-the-art machine learning technique to segment with the SVM on the burn images. The proposed method is an Extended Model of Segmentation and SVM (EMS-SVM).  The model is trained and tested on the COCO and BIS data set respectively. Finally, the classification strategy of the model is really advantageous for the burn classification.