A novel adaptive LBP-based descriptor for color image retrieval
作者：
Highlights：
• Two new local feature extraction methods are proposed for color image retrieval.
• The methods are based on Radial Mean Completed Local Binary Pattern method.
• A new similarity measure is proposed for comparing the prototypes of images.
• Particle Swarm Optimization is used for feature weighting to improve the accuracy.
• The results show the merit of the proposed methods comparing with other methods.
摘要
•Two new local feature extraction methods are proposed for color image retrieval.•The methods are based on Radial Mean Completed Local Binary Pattern method.•A new similarity measure is proposed for comparing the prototypes of images.•Particle Swarm Optimization is used for feature weighting to improve the accuracy.•The results show the merit of the proposed methods comparing with other methods.
论文关键词：Image retrieval,Color Radial Mean Completed Local Binary Pattern (CRMCLBP),Prototype Data Model (PDM),Particle Swarm Optimization (PSO),Feature weighting,K-means clustering
论文评审过程：Received 6 November 2018, Revised 11 March 2019, Accepted 12 March 2019, Available online 12 March 2019, Version of Record 19 March 2019.
论文官网地址：https://doi.org/10.1016/j.eswa.2019.03.020
