CAJ | 학술논문

脱机手写数字体自动识别一直以来是图像处理与模式识别领域中的研究热点及具有较高的实用价值。文章对手写数字体研究现状进行概述，着重分析了预处理中的二值化、细化等方法，对特征提取中较重要的统计特征和结构特征分别进行阐述，讨论了分类识别中BP神经网络、支持向量机等关键技术的各种主流方法。通过分析，以期使读者对手写数字体识别的进展有较全面的了解，并对未来的研发有切实的帮助。
탈궤수사수자체자동식별일직이래시도상처리여모식식별영역중적연구열점급구유교고적실용개치。문장대수사수자체연구현상진행개술，착중분석료예처리중적이치화、세화등방법，대특정제취중교중요적통계특정화결구특정분별진행천술，토론료분류식별중BP신경망락、지지향량궤등관건기술적각충주류방법。통과분석，이기사독자대수사수자체식별적진전유교전면적료해，병대미래적연발유절실적방조。
Handwritten Digital Automatic Recognition has been a very active research topic in the field of image processing and pattern recognition and is of great practical importance. This paper gives an overview of the current research on handwritten digital identification, focusing on the analysis of binarization and thinning in the preprocessing and on the illustration of the more important statistical and structural features in feature extraction, and discusses various main methods of key technologies in classification and recognition such as the BP neural network and the support vector machine. The goal of this paper is to provide a more comprehensive understanding of handwritten digital identification for future research and development.