• Sumit Chopra
• Dr. V . K . Banga

Breast cancer is one of the deadliest problems that the women in European countries are facing. Several computer detection methodologies have been proposed which help experts to identify the suspicious region that are difficult to find with the naked eye, thus helping in diagnosis and detection of cancer. A methodology for classification of the regions extracted from mammogram to classify them as mass or non-mass is proposed. Mammographic Image Analysis Society (MIAS) database was used for image acquisition. Two new features were described for texture of the region of interest which has added advantage that even if the region of interest is not segmented properly, then the results will not be affected as texture is invariant to the shape of the region of interest extracted. Internal and external masks were used for analysis of texture. Support vector machine for used for classification of region as masses or non-masses. The results showed a large improvement as compared with the existing state of art techniques.

Content Based Image Retrieval (CBIR) , Masses, Classification, Diversity tree, Cladogram

"Content based image retrieval system for Classification of breast region as mass or non masses using new shape invariant texture feature and SVM.", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.5, Issue 2, page no.74-79, February-2018, Available :http://www.jetir.org/papers/JETIR1802010.pdf

"Content based image retrieval system for Classification of breast region as mass or non masses using new shape invariant texture feature and SVM.", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.5, Issue 2, page no. pp74-79, February-2018, Available at : http://www.jetir.org/papers/JETIR1802010.pdf