An Improved Tumour Identification with Gabor Wavelet Segmentation

Research work on medical images is always challenging because this is linked with the lives of human. Results obtained with medical images will help in the early detection and diagnosis diseases, like cancer, and later for diagnosis of those images. In cancer, breast cancer is a major cause of death in women. Breast tumor is harmful and begins in the cells of the breast. In our proposed work we are working on improving the segmentation part to help in early detection of breast cancer. It can be performed using techniques like Preprocessing, feature extraction and classification. Here we used Gabor wavelet feature extraction technique for detecting breast tumor areas which divides image into different slices. This way of dividing image into slices helps us to make comparison between lesion area and normal area. So at last component vectors are made out of the pixel's power and the Gabor wavelet reactions measured at various scales. Here we used data set of 320 images of tumor affected breast and Gabor feature extraction method applied on it. Later, SVM (Support Vector Machine) is used for classification purpose.