Digital mammogram images are generally used in medical field as a standard tool for enhancing, transmission and restoring of data. The procedure of image processing is applied to diagnose breast cancer from mammographic ROI image. The quality of mammogram pictures are very low and are sometimes influenced by X-Ray absorption properties of an anatomic parts, size as well as shape. The method of pre-processing help to enhance the raw mammogram image obtained from sensors to aid in an identification of tumors. The proposed work uses Discrete Wavelet Transform (DWT) to decompose the given grayscale image. The textual and statistical features are been extracted from spatial domain coefficients along with frequency domain coefficients. The feature extraction method used in this work is Gray-Level Cooccurrence Matrix (GLCM). Classification of image is performed using support vector and artificial neural network as benign or malignant. The proposed method is applied on Mammographic Image Analysis Society (MIAS) database. The images of the database have to undergo training, testing and validation stages.
Soumya Hundekar, Saritha Chakrasali
BNM Institute of Technology, India
Mammography, Feature Extraction, Support Vector Machine, Artificial Neural Network, Gray-Level Co-occurrence Matrix, Mammographic Image Analysis Society, Discrete Wavelet Transform | Published By :
ICTACT
Published In :
ICTACT Journal on Soft Computing
( Volume: 9 , Issue: 2 )
Date of Publication :
Januray 2019
|