AN INTELLIGENT CONTENT BASED IMAGE RETRIEVAL SYSTEM FOR MAMMOGRAM IMAGE ANALYSIS

Journal of Engineering Science and Technology

T. S. SUBASHINI¹  K. VAIDEHI¹ 

[1] Department of Computer Science and Engineering, Annamalai University, Annamalai Nagar, 608002, Tamil Nadu, India ;

关键词: Content-based image retrieval;    Computer-aided diagnosis;    Supportvector machine;    Statistical descriptors;    Breast density;   

DOI  :  

来源: DOAJ

【 摘 要 】

An automated segmentation method which dynamically selects the parenchymal region of interest (ROI) based on the patients breast size is proposed from which, statistical features are derived. SVM classifier is used to model the derived features to classify the breast tissue as dense, glandular and fatty. Then K-nn with different distance metrics namely city-block, Euclidean and Chebchev is used to retrieve the first k similar images closest to the given query image. The proposed method was tested with MIAS database and achieves an average precision of 86.15%. The results reveals that the proposed method could be employed for effective content based mammograms retrieval.

【 授权许可】

Unknown