Scalding is one of the major accidents and is often life threatening. The research attempts to find automated solution for classifying scalding (grade 1, grade 2, and grade 3). In India the statistics show that more than 50% children are affected by scalds and thermal burn. The research finds an automated solution for classifying scald burn as superficial, partial thickness and full thickness scald. The development and implementation of the proposed work are of significant importance specifically in rural areas where medical facilities are scarce. A scalding burn image database is formed with images collected from hospitals and other open sources. The pattern analysis or pattern classifier technique namely Support Vector Machine (SVM) and k-Nearest Neighbors algorithm (KNN) is used in this work. SVM is found to give best results in comparison with the KNN classification.
[1]
Begoña Acha,et al.
CAD Tool for Burn Diagnosis
,
2003,
IPMI.
[2]
Begoña Acha,et al.
Segmentation of burn images based on color and texture information
,
2003,
SPIE Medical Imaging.
[3]
Malini suvarna,et al.
Performance Measure and Efficiency of Chemical Skin Burn Classification Using KNN Method
,
2015
.
[4]
Vladimir N. Vapnik,et al.
The Nature of Statistical Learning Theory
,
2000,
Statistics for Engineering and Information Science.
[5]
Begoña Acha,et al.
Classification of burn wounds using support vector machines
,
2004,
SPIE Medical Imaging.
[6]
U C Niranjan,et al.
Diagnosis of Burn Images using Template Matching, k-Nearest Neighbor and Artificial Neural Network
,
2013
.