الفهرس | Only 14 pages are availabe for public view |
Abstract Breast cancer is that the foremost visit cancer among ladies, influencing 2.1 million women every year, and moreover causes the most prominent range of cancer-related passing among women. The point of this thesis is to utilize an Ant Colony-based classification framework to extricate a set of rough rules for classification of breast cancer images into two classes: benign or malignant named ACO-RCS which has modern characteristics that build it distinctive from the existing strategies that have utilized the Ant Colony Optimization (ACO) for classification tasks. Digital mammograms have end up the best strategies for the detection of breast most cancers. The purpose of this studies is to growth the diagnostic accuracy for maximum class among malignant and benign abnormalities in digital mammograms with the aid of decreasing the variety of misclassified cancers. The strategy is connected to genuine mammogram pictures are collected from ‘Mammographic Image Analysis Society’ (MIAS) database. Ordinarily, the classification system begins with the preprocessing which incorporates digitization of the mammograms with totally different sampling and quantization, Adaptive Middle Channel is utilized as a pre-processing method to decrease commotion in an image. Then the ROI chosen from the digitized mammogram are improved. The segmentation method is intended to find suspicious regions and to extricate the region of interest (ROI) from the preprocessed pictures that are segmented by using Gaussian mixture model (GMM) technique. |