Compact MQDF classifiers using sparse coding for handwritten Chinese character recognition. (April 2018)

• Record Type: Journal Article
• Authors: Wei, Xiaohua; Lu, Shujing; Lu, Yue
• Is Part Of: Pattern recognition. Volume 76(2018:Apr.)
• Page Start: 679
• Page End: 690
• Publication Date: 2018-04
• Subjects: Sparse coding -- Compact MQDF classifier -- Multiple dictionary learning -- Handwritten Chinese character recognition; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2017.09.044
• Languages: English
• ISSNs: 0031-3203

Abstract

Highlights: We use sparse coding to compact the parameters of MQDF classifier. An analysis is given to indicate that using the sparse representation to build compact MQDF classifier is feasible and valid. We learn multiple dictionaries rather than single dictionary to reduce the computational complexity. A weight-based assignment strategy is proposed to further alleviate the degradation of recognition accuracy. Experiments demonstrate the validation of the proposed methods. Abstract: The modified quadratic discriminant function (MQDF) is an effective classifier for handwritten Chinese character recognition (HCCR). However, it suffers from high memory requirement for the storage of its parameters, which makes it impractical to be embedded in memory limited hand-held devices. In this paper, we explore the applicability of sparse coding to build compact MQDF classifiers. To be specific, we use sparse coding to compact the parameters of MQDF. Two methods of sparse coding, viz., the maximum likelihood-based method and the K-SVD method, are adopted to build two compact MQDF classifiers, namely, MQDF-ML classifier and MQDF-KSVD classifier. Furthermore, we learn multiple dictionaries rather than single dictionary for sparse coding, because the multiple dictionary learning is capable of not only greatly reducing the computational complexity, but also alleviating the degradation of recognition accuracy, compared to the single dictionary learning. Experiments and comparison with the existing method have demonstrated the effectiveness of our proposed method for the issue of unconstrained handwritten Chinese character recognition.