A Nonnegative Locally Linear KNN model for image recognition

作者：

Highlights：

• Non-zero representation coefficients of Locally Linear KNN (LLKNN) are investigated.

• A Nonnegative Locally Linear KNN (NLLKNN) model is proposed for image recognition.

• A multiplicative iterative algorithm with proof of convergence is proposed.

• The robustness of NLLKNNC and NLLNMC to noises and occlusions are investigated.

• Extensive experiments show the superiority of the proposed method.

摘要

•Non-zero representation coefficients of Locally Linear KNN (LLKNN) are investigated.•A Nonnegative Locally Linear KNN (NLLKNN) model is proposed for image recognition.•A multiplicative iterative algorithm with proof of convergence is proposed.•The robustness of NLLKNNC and NLLNMC to noises and occlusions are investigated.•Extensive experiments show the superiority of the proposed method.

论文关键词：Sparse minimization,Nonnegative,Locally linear KNN,Robustness,Image recognition

论文评审过程：Received 12 May 2017, Revised 7 May 2018, Accepted 21 May 2018, Available online 22 May 2018, Version of Record 26 May 2018.

论文官网地址：https://doi.org/10.1016/j.patcog.2018.05.024