Description method of Illumination invariant image features
作者：
Highlights：
• We propose an improved pattern named as weighted local synergy gradient pattern (WLSGP).
• Anisotropic differential synergy excitation pattern (ADSEP) is proposed to solve the directional difference problem of the original WLD operator.
• Based on WLSGP and ADSEP, A novel illumination–insensitive feature descriptor called as AWSDG is proposed.
• Abundant experiments under different illumination variations validate the effectiveness of proposed methods.
摘要
•We propose an improved pattern named as weighted local synergy gradient pattern (WLSGP).•Anisotropic differential synergy excitation pattern (ADSEP) is proposed to solve the directional difference problem of the original WLD operator.•Based on WLSGP and ADSEP, A novel illumination–insensitive feature descriptor called as AWSDG is proposed.•Abundant experiments under different illumination variations validate the effectiveness of proposed methods.
论文关键词：Feature extraction,Illumination variation,LBP,LGP,WLD
论文评审过程：Received 8 August 2019, Revised 27 April 2020, Accepted 27 April 2020, Available online 29 April 2020, Version of Record 5 May 2020.
论文官网地址：https://doi.org/10.1016/j.image.2020.115870
