Multiple Color Channel Local Extrema Patterns for Image Retrieval
A novel feature descriptor, multiple color channel local extremal pattern (MCLEP) is proposed in this manuscript. MCLEP combines the key features of local binary and local quantized extrema patterns in a specified neighborhood. Multi-distance information computed by the MCLEP aids in robust extraction of the texture arrangement. Further, MCLEP features are extracted for each color channel of an image. Retrieval performance of the MDLP is evaluated on benchmark datasets for CBIR, namely Corel-5000, Corel-10000, and MIT-Color Vistex, respectively. Proposed technique exhibits substantial improvement as compared to other recent feature descriptors in terms of ARP and ARR on standard databases.
KeywordsColor Texture LBP, LQEP, and retrieval
- 2.Krizhevsky A, Sutskever I, Hinton GE (2012) Imagenet classification with deep convolutional neural networks. In: Advances in neural information processing systems, pp 1097–1105Google Scholar
- 4.Tsikrika T, Popescu A, Kludas J (2011) Overview of the Wikipedia image retrieval task at ImageCLEF 2011. In: The working notes for the CLEF 2011 labs and workshop, Amsterdam, The Netherlands, 19–22 Sept 2011Google Scholar
- 11.Koteswara Rao L, Venkata Rao D, Pratap Reddy L (2016) Local mesh quantized extrema patterns for image retrieval. SpringerPlus 5:976. https://doi.org/10.1186/s40064-016-2664-9
- 12.Koteswara Rao L et al (2018) Local color oppugnant quantized extrema patterns for content based natural and texture image retrieval. In: Multidimensional systems and signal processing. Springer (accepted)Google Scholar
- 13.Koteswara Rao L, Venkata Rao D (2015) Local quantized extrema patterns for content-based natural and texture image retrieval. Hum Centric Comput Inf Sci. https://doi.org/10.1186/s13673-015-0044-z