Application of image processing algorithms on medical images is always challenging. This always help in early detection stages and later for diagnosis of different cancer issues. As breast cancer being second major cause of death in women, so our interest is on this topic to improve the accuracy to segment the tumors in mammograms. For this purpose in our paper, we present algorithm based on region and contour based and some clustering segmentation techniques to recognize tumors in breast images. Our first step in this algorithm require the level set. We have used Spatial Fuzzy Clustering (SFC) to improve region growing. In Second step, Artificial Neural Network is used to regulate all level set parameters. This approach is used with Genetic Algorithm. Third step is of feature extraction. This step is used to extract features from segmented images to train a classified to determine whether a tumour in label as normal or tumour area. As last step, classification algorithms are used to make difference between tumours affected image or normal image. We will test with ANN algorithms. © 2016, International Journal of Pharmacy and Technology. All rights reserved.