The objective of the project is to classify skin lesion and cancer. A novel method is proposed that combines color and texture for the segmentation of skin lesions from unaffected skin region in an image. This project proposes a novel approach for classification of skin lesion and cancer images. The proposed work comprises of Pre-Processing, Segmentation, Feature extraction and Classification. In the Pre- Processing stage, Anisotropic diffusion Filter is implemented to remove noise and undesired structures from the images. In the Segmentation stage Fast Fuzzy C Means clustering method is implemented in order to acquire a contour by means of the gradient flow that minimizes an energy function with a distance regularization term and an external energy that drives the motion of the zero level set toward desired locations. The Gray level Co-occurrence Matrix (GLCM) and bandlet transform are used to estimate the features of the segmented image. The convolutional neural network classifier is employed for the classification task, utilizing feature vectors derived from gray level co- occurrence (GLCM) features. The classification results are evaluated with the use of accuracy, sensitivity and specificity. An automated Matlab tool is developed for classification of skin lesion and cancer.
Key words: Dermoscopic Digital images, Feature extraction, gray level co -occurrence matrix and bandlet transform
|
