Taking advantage of multi-regions-based diagonal texture structure descriptor for image retrieval

作者：

Highlights：

• A novel diagonal texture structure descriptor is proposed.

• A new framework is proposed to generate the feature vector of an image.

• Hue and saturation are quantized for detailed description of color difference.

• We segment image into multi-regions for making more accurate content match.

摘要

•A novel diagonal texture structure descriptor is proposed.•A new framework is proposed to generate the feature vector of an image.•Hue and saturation are quantized for detailed description of color difference.•We segment image into multi-regions for making more accurate content match.

论文关键词：Diagonal texture structure descriptor,Multi-regions,Receptive field,Visual system,Content-based image retrieval

论文评审过程：Received 22 April 2016, Revised 2 November 2017, Accepted 3 December 2017, Available online 6 December 2017, Version of Record 22 December 2017.

论文官网地址：https://doi.org/10.1016/j.eswa.2017.12.006