Improved search space shrinking for medical image retrieval using capsule architecture and decision fusion

作者：

Highlights：

• A capsule classification system which significantly improves the top-1 accuracy.

• A decision fusion using RBC and W-DCT at the second step to refine the accuracy.

• IRMA error of 124.25 is obtained using 193 class-code categories.

摘要

•A capsule classification system which significantly improves the top-1 accuracy.•A decision fusion using RBC and W-DCT at the second step to refine the accuracy.•IRMA error of 124.25 is obtained using 193 class-code categories.

论文关键词：Content-based image retrieval,Machine learning,Image processing,Image retrieval

论文评审过程：Received 18 March 2020, Revised 12 October 2020, Accepted 25 December 2020, Available online 30 December 2020, Version of Record 17 January 2021.

论文官网地址：https://doi.org/10.1016/j.eswa.2020.114543