An Efficient Mammogram Image Retrieval System Using an Optimized Classifier

The computerized examination of mammograms in the breast cancer prevention is gaining much importance. The paper which is introduced proposes an adequate mammogram image retrieval methodology utilizing the optimized classifier. At first, the info mammogram image is brought as of the database and it is then pre-handled by utilizing the Modified Weiner. This filtered image undergoes the pectoral removal. This is followed with the method of Feature extraction. The extorted features are then categorized to 3 classes namely benign, malignant and normal by utilizing an optimized classifier. The ‘Modified Adaptive Neuro-Fuzzy Inference System’ is optimized using the ABC Algorithm (‘Artificial Bee Colony’). Score values of such classified images are determined utilizing the ‘Principal Component Analysis’. Then repeat a similar process for query images and finally, the minimal distance is evaluated betwixt these 2 images This is finished using the Euclidian distance and in this way the image having less distance on diverged from the question is recovered. The proposed mammogram image retrieval methodology is implemented on the stage called MATLAB and it is evaluated by utilizing disparate database images.