Content-Based Image retrieval (CBIR) is an application of computer vision which tries to cope up with the hindrance of image retrieval i.e. searching of an image from the large repositories. In CBIR, images are denoted as features in the dataset. The two key parameters involved in retrieving the actual content includes the extracting of image features and features matching. Once the image features are extracted, the homogeneous attributes are measured with the help of similarity metrics. Similarity matrix is one of the key issues in CBIR. Their exits number of similarity matrix. Most popularly used similarity metrics are described and evaluated in this paper. They are evaluated on standard Wang Image dataset which is considered as a benchmark database for CBIR. Result shows that Chebychev Metrics gives the highest average accuracy of 92.56% whereas in respect of response time, Euclidean Metrics gives the best time performance with average elapsed time of 1.275573 seconds.
Evaluation Of Distance Measures In Content Based Image Retrieval
2019-06-01
3339209 byte
Conference paper
Electronic Resource
English
Content-based image retrieval with the normalized information distance
British Library Online Contents | 2008
|Evaluation of feature measures and similarity measures for content-based retrieval [2606-25]
British Library Conference Proceedings | 1995
|Here's Waldo: Content Based Image Retrieval
British Library Online Contents | 1998
|A content-based image retrieval system
British Library Online Contents | 1998
|Learning distance functions for image retrieval
IEEE | 2004
|