Image Retrieval Using Spatial Multi-color Coherence Vectors Mixing Location Information
Citations Over TimeTop 16% of 2008 papers
Abstract
Color coherence vectors (CCV) are commonly used for image retrieval in content-based image retrieval (CBIR). It is a more sophisticated form of histogram refinement, in which histogram buckets are partitioned based on spatial coherence, but a single CCV is not enough for efficient and robust image retrieval system. In this paper, a method for image retrieval using the multi-CCV mixing location is proposed. In this method, we separate the image into core region and minor region, adopt hash table to store the pixel into the appropriate color of our color set, and mixing location information with the traditional method to bring multi-CCV of the image. The multi-CCV mixing location gives much better retrieval efficiency, but with higher computations and appear to perform better than the traditional CCV.
Related Papers
- → Expanding histogram of colors with gridding to improve tracking accuracy(2015)23 cited
- → <title>Bounds for the discrimination power of color indexing techniques</title>(1994)26 cited
- → Uniform Extended Local Ternary Pattern for Content Based Image Retrieval(2018)1 cited
- Scene change detection algorithm on specific movie(2009)
- CBIR using Speech, Text & Image Query for Mobile Device(2016)