HEp-2 cells classification via sparse representation of textural features fused into dissimilarity space
作者：
Highlights：
• A method for HEp-2 cells classification based on fluorescence staining patterns is proposed.
• Gradient and textural features are captured using two descriptors.
• The descriptors are fused into the dissimilarity space.
• A sparse representation-based classification scheme undertakes the final classification.
• Classification accuracy up to 75.1% in cell level and 85.7% in image level is obtained.
摘要
•A method for HEp-2 cells classification based on fluorescence staining patterns is proposed.•Gradient and textural features are captured using two descriptors.•The descriptors are fused into the dissimilarity space.•A sparse representation-based classification scheme undertakes the final classification.•Classification accuracy up to 75.1% in cell level and 85.7% in image level is obtained.
论文关键词：HEp-2 cells,Staining patterns classification,Local binary patterns,SIFT descriptors,Dissimilarity fusion,Dissimilarity representation,Sparse representation,Multiple-level representation
论文评审过程：Available online 1 October 2013.
论文官网地址：https://doi.org/10.1016/j.patcog.2013.09.026
