Main Article Content
The Oral Epithelial Dysplasia (OED) lesion is referred as a pre-cancerous lesion. It is a stepwise growth to cancer within the oral mucosa. The primary occurrence of a pre-cancer lesion is consequently increases the growth of the cancer cells in its surrounding area. In the proposed work, the Data Wavelet Transformation is applied to discriminate the normal and oral epithelial dysplasia disease affected images. For this the microscopic images have collected from Raja Muthiah Dental College and Hospital. The two feature extraction techniques namely Histogram Oriented features and Local Binary Pattern are used to extract the features. The extracted features are given as the input to Back Propagation Neural Network. The histogram oriented features with Back Propagation neural network gives the satisfactory results of 85%.