A software system for automated identification and retrieval of moth images based on wing attributes

作者：

Highlights：

• A species identification and retrieval system for moth images.

• Semantically related visual (SRV) attributes and signatures.

• An algorithm to detect the co-occurrence patterns of SRV attributes.

• Attribute signature based image identification and retrieval framework.

• Performance improvement with SRV attributes and their co-occurrence patterns.

摘要

Highlights•A species identification and retrieval system for moth images.•Semantically related visual (SRV) attributes and signatures.•An algorithm to detect the co-occurrence patterns of SRV attributes.•Attribute signature based image identification and retrieval framework.•Performance improvement with SRV attributes and their co-occurrence patterns.

论文关键词：Entomological image identification and retrieval,Semantically related visual attributes,Attribute co-occurrence pattern detection

论文评审过程：Received 23 July 2013, Revised 12 September 2015, Accepted 14 September 2015, Available online 28 September 2015, Version of Record 27 November 2015.

论文官网地址：https://doi.org/10.1016/j.patcog.2015.09.012