Analysis of Different Filtering Methods for Preprocessing and GLCM Feature Extraction Using Wavelet in Mammogram Images

Breast cancer is a stand-out surrounded by the most widely perceived diseases and has a high rate of
mortality around the world, significantly risking the health of the females. Among existing all modalities
of medical scans, mammography is the most preferred modality for preliminary examination of breast
cancer. To assist radiologists, a computer-aided diagnosis (CAD) is enhancing and important medical
systems for mammographic lesion analysis. In mammogram images, micro-calcifications are one of
the imperative signs for breast cancer detection. Mammographic medical scan may present unwanted
noise and CAD systems are very sensitive to noise. Early stage detection for any medical image
analysis application like brain tumor detection, breast cancer detection is considered as an important
step. Micro calcification is small calcium deposits in the breast region and mammogram images are of
low contrast. Thus, in this work, different types of filtering techniques used for noise reduction and
image enhancement for medical image processing are analyzed on mini-MIAS mammogram image
databases. Anisotropic diffusion with wavelet filtering method shows best results for enhancement
and noise removal of the image. This filtered image is segmented; region of interest (ROI) is extracted
through global Thresholding technique with discrete wavelet transform (DWT). Gray level cooccurrence
matrix is used to extract the important features. Here, seven features are extracted for
different categories of micro calcified images like normal, benign, malignant. Results show that from
extracted features the values of malignant and micro calcified images are same whereas normal and
benign are same. This proposed methodology can help to categorize different classes of images.