Deep-seated features histogram: A novel image retrieval method

作者：

Highlights：

• Low-level features are extracted by simulating the human orientation selection and color perception mechanisms.

• Ranking whitening is proposed for extracting deep features via low-level features and reasonably combining them to obtain deep-seated features.

• The proposed method is straightforward and reduces the vector dimensionality.

• Deep-seated features can describe image contents in terms of colors and edge orientations and identify similar scene styles.

摘要

•Low-level features are extracted by simulating the human orientation selection and color perception mechanisms.•Ranking whitening is proposed for extracting deep features via low-level features and reasonably combining them to obtain deep-seated features.•The proposed method is straightforward and reduces the vector dimensionality.•Deep-seated features can describe image contents in terms of colors and edge orientations and identify similar scene styles.

论文关键词：Image retrieval,VGG-16 network,orientation selection,color perception,deep-seated features histogram

论文评审过程：Received 3 May 2020, Revised 15 December 2020, Accepted 1 March 2021, Available online 6 March 2021, Version of Record 13 April 2021.

论文官网地址：https://doi.org/10.1016/j.patcog.2021.107926