Improved adaptive image retrieval with the use of shadowed sets

作者：

Highlights：

• The underlying originality of the proposed method relies on the application of the shadowed sets theory in image retrieval, which is a first involving shadowed sets in category of problems.

• Techniques of automatic parameter selection are developed for automatically obtaining shadowed regions by combining saliency detection and edge detection.

• A three-way division approach of images is presented to automatically split an image into three regions: salient regions, non-salient regions, and shadowed regions.

• The salient regions and the shadowed regions are jointly used as the correctly detected regions to improve performance of segmentation and accuracy of retrieval.

摘要

•The underlying originality of the proposed method relies on the application of the shadowed sets theory in image retrieval, which is a first involving shadowed sets in category of problems.•Techniques of automatic parameter selection are developed for automatically obtaining shadowed regions by combining saliency detection and edge detection.•A three-way division approach of images is presented to automatically split an image into three regions: salient regions, non-salient regions, and shadowed regions.•The salient regions and the shadowed regions are jointly used as the correctly detected regions to improve performance of segmentation and accuracy of retrieval.

论文关键词：Shadowed sets,Saliency detection,Image retrieval,Image segmentation,Edge detection

论文评审过程：Received 8 May 2018, Revised 6 January 2019, Accepted 25 January 2019, Available online 28 January 2019, Version of Record 13 February 2019.

论文官网地址：https://doi.org/10.1016/j.patcog.2019.01.029