文章基本信息

标题：Sketch-based Image Retrieval Using SRA Saliency
本地全文：下载
作者：Safiya P.C ; Sandhya V ; Niyas N 等
期刊名称：International Journal of Innovative Research in Computer and Communication Engineering
印刷版ISSN：2320-9798
电子版ISSN：2320-9801
出版年度：2017
卷号：5
期号：3
页码：5573
DOI：10.15680/IJIRCCE.2017.0503121
出版社：S&S Publications
摘要：Softwaresystems which uses hand drawn sketches as the queryfor image retrieval are known as sketchbased image retrieval (SBIR) systems. One of the mainadvantages of these type of systems is that it can take sketchesas input . Main task in SBIR is to measure the similaritybetween a query sketch and an image. As a solution tothisproblem, we propose a Sketch Based Image Retrieval approach using SRA Saliency which utilizes global as well aslocal image features. Here we propose two types of contour maps namely global contour map (GCM) and the salientcontour map (SCM). SCM is to identify objects inside images which matches with the query sketch. Apart from thesecontour maps we proposean image feature called angular radial partitioning (ARP) feature combined with angularradial orientation partitioning (AROP) feature. AROP-ARP feature fully utilizes the edge pixel’s informationin GCMand SCM to identify the feature vectors. SBIR systems are very effective in the field of medical diagnosis, photosharinglibrary etc.
关键词：SBIR; Salient Contour; Contour Reinforcement; AROP feature; SRA Saliency