摘要
为了古典文献文字识别效果更好,在分析前人研究成果的基础上,基于多学科交叉,对遗传算法进行改进:建立经过优化的初始化种群,为交叉选择方法提供多样性的信息,利用柯西变异与高斯变异结合形成PM生成器.实验表明:该技术有效地提高了求解质量,较好地优化算法性能.
An Improvement method of Genetic Algorithms Based on Multidisciplinary Intersection was introduced:for the better recognition effect of classical documents word,based on the analysis of previous research results,establishment of optimized initialization population,providing diverse information for cross-selection methods,formation of PM Generator by combining Cauchy variation with Gauss variation.Experiments showed that this technique could effectively improve the quality of solution and the optimization performance of the algorithm.
作者
张敬花
马海云
张忠林
ZHANG Jing-hua;MA Hai-yun;ZHANG Zhong-lin(School of Marxism,Tianshui Normal University,Gansu Tianshui 741001,China;School of electronic information and electrical engineering,Tianshui Normal University,Gansu Tianshui 741001,China;School of Electronic and Information Engineering,Lanzhou Jiaotong University,Lanzhou 730070,China)
出处
《中央民族大学学报（自然科学版）》
2019年第3期56-60,共5页
Journal of Minzu University of China(Natural Sciences Edition)
基金
甘肃省自然科学基金项目(No.18JR3RE245)
教育部人文社会科学研究规划基础(No.14YJA870014)
关键词
古典文献汉字识别
属性特征
改进遗传算法
小波变换
分治策略
classical Chinese character recognition
attribute features
improved genetic algorithm
wavelet transform
divide and conquer strategy
