Subclass Discriminant Analysis of morphological and textural features for HEp-2 staining pattern classification
作者：
Highlights：
• We propose a HEp-2 classification strategy based on Subclass Discriminant Analysis.
• Our approach copes well with the high within-class variance of HEp-2 patterns.
• We investigate the individual and combined contribution of different image attributes.
• We integrate morphological, global and local textural information.
• Our approach provides staining pattern classification accuracy of about 90%.
摘要
Highlights•We propose a HEp-2 classification strategy based on Subclass Discriminant Analysis.•Our approach copes well with the high within-class variance of HEp-2 patterns.•We investigate the individual and combined contribution of different image attributes.•We integrate morphological, global and local textural information.•Our approach provides staining pattern classification accuracy of about 90%.
论文关键词：Indirect Immunofluorescence,Fluorescence pattern classification,Subclass Discriminant Analysis,Morphological analysis,Textural analysis
论文评审过程：Available online 1 October 2013.
论文官网地址：https://doi.org/10.1016/j.patcog.2013.09.024
