文章基本信息

标题：Performance evaluation of the relevance feedback and Histogram of Oriented Gradients (HOG) based CBIR techniques – A Review
本地全文：下载
作者：Tania Gupta ; Dr. Amandeep Verma
期刊名称：International Journal of Computer Trends and Technology
电子版ISSN：2231-2803
出版年度：2015
卷号：23
期号：2
页码：49-52
DOI：10.14445/22312803/IJCTT-V23P111
出版社：Seventh Sense Research Group
摘要：There are several Content Bases Image Retrieval (CBIR) techniques accessible round the globe within the IT space. The CBIR techniques are used to fetch the similar images on the premises of varied high level or low level image features such as color, shape, pattern, texture, histogram etc. These CBIR techniques use different image features for the purpose of similarity evaluation. The CBIR techniques may be either based upon the usual one time image results or user feedback or relevance feedback for the image results optimization according to the users recommendations on the irrelevant images. The various CBIR techniques are being evaluated in this paper in order to find the appropriate techniques for our literature survey and implementation. The CBIR techniques that embody the relevance feedback based upon fuzzy with semantic memory, standard type or relevance feedback, LBPHOG descriptor and Gradient field HOG descriptor methods have been evaluated on the premise of accuracy, speed and various other factors in this paper. The results have been evaluated using the various feature model, relevance model, accuracy and other performance parameters.
关键词：CBIR; performance evaluation; relevance feedback; HOG; etc.