Multi-layer discriminative dictionary learning with locality constraint for image classification
作者：
Highlights：
• A powerful architecture, called the multi-layer discriminative dictionary learning (MDDL) with locality constraint, is proposed for image classification.
• Through the multi-layer dictionary learning, the robust dictionary is obtained in the final layer, where the separability of coding vectors from different classes is also increased.
• Benefiting from joint classifier training and multi-layer dictionary learning, the discriminability of the learned coding vectors is further enhanced.
• By utilizing the graph Laplacian matrices based on the learned dictionaries, not only the locality information of the original data is preserved, but also it can avoid very large values in the coding vectors to reduce the test error caused by overfitting.
摘要
•A powerful architecture, called the multi-layer discriminative dictionary learning (MDDL) with locality constraint, is proposed for image classification.•Through the multi-layer dictionary learning, the robust dictionary is obtained in the final layer, where the separability of coding vectors from different classes is also increased.•Benefiting from joint classifier training and multi-layer dictionary learning, the discriminability of the learned coding vectors is further enhanced.•By utilizing the graph Laplacian matrices based on the learned dictionaries, not only the locality information of the original data is preserved, but also it can avoid very large values in the coding vectors to reduce the test error caused by overfitting.
论文关键词：Multi-layer discriminative dictionary learning,Locality constraint,Classifier training,Image classification
论文评审过程：Received 29 October 2017, Revised 30 August 2018, Accepted 19 February 2019, Available online 19 February 2019, Version of Record 25 February 2019.
论文官网地址：https://doi.org/10.1016/j.patcog.2019.02.018
