CAJ | 학술논문

文章设计了一种基于粗糙集理论的手写体汉字识别决策系统,依据并行遗传算法在处理大型数据库上的特有优势,将信息熵理论作为粗糙集不确定性的度量,给出一种约简冗余条件属性求解决策信息系统约简集的多群体并行遗传算法;并提出了一种基于规则置信度的手写体汉字识别规则融合算法,提高识别准确率,对SCUT_IRAC手写体汉字库中手写体汉字进行识别实验,结果证明,该系统具有较强的手写体汉字识别能力.
문장설계료일충기우조조집이론적수사체한자식별결책계통,의거병행유전산법재처리대형수거고상적특유우세,장신식적이론작위조조집불학정성적도량,급출일충약간용여조건속성구해결책신식계통약간집적다군체병행유전산법;병제출료일충기우규칙치신도적수사체한자식별규칙융합산법,제고식별준학솔,대SCUT_IRAC수사체한자고중수사체한자진행식별실험,결과증명,해계통구유교강적수사체한자식별능력.
A rough set-based recognition decision system is constructed.The entropy theory is used to measure the uncertainties of rough sets.By taking the advantages of the parallel genetic algorithm in dealing with huge data sets,a multi-population genetic algorithm is presented to reduce the overlapped attribute of the decision information system to get the reduced set.A kind of rule fusion method for Chinese character recognition based on confidence is designed to enhance the rate of accuracy.Hand-written Chinese characters from SCUT-IRAC HCCLIB are tested and the results confirm that the proposed method has the ability of recognizing handwritten Chinese characters.