In-air handwritten Chinese character recognition with locality-sensitive sparse representation toward optimized prototype classifier. (June 2018)

• Record Type: Journal Article
• Authors: Qu, Xiwen; Wang, Weiqiang; Lu, Ke; Zhou, Jianshe
• Is Part Of: Pattern recognition. Volume 78(2018:Jun.)
• Page Start: 267
• Page End: 276
• Publication Date: 2018-06
• Subjects: Locality-sensitive sparse representation based classification -- In-air handwritten Chinese character recognition -- Minimum classification error -- Handwritten Chinese character recognition; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2018.01.021
• Languages: English
• ISSNs: 0031-3203

Abstract

Highlights: We present a new classifier, named LSROPC. We apply the LSROPC to in-air handwritten Chinese character recognition (IAHCCR). We built the IAHCC-UCAS2016 dataset for IAHCCR. The LSROPC can be applied to other fields. Abstract: Locality-sensitive sparse representation based classification has been shown to be effective for in-air handwritten Chinese character recognition (IAHCCR). In this paper, we present a locality-sensitive sparse representation toward optimized prototype classifier (LSROPC) for IAHCCR. In the proposed LSROPC, the learned dictionary can not only preserve local data structures, but also require the reconstruction of a pattern to get as close as possible to the prototype optimized by the minimum classification error (MCE) approach. So the LSROPC can help improve the classification accuracy effectively. The experiments are carried out on the datasets of traditional handwritten Chinese characters and in-air handwritten Chinese characters and the datasets designed for face recognition. The experimental results demonstrate the effectiveness of the proposed method.