您的位置:首页  > 论文页面

基于自适应遗传算法的血细胞分子光谱图像波段选择

发表时间:2009-09-30  浏览量:1266  下载量:593
全部作者: 戴春妮,李庆利,刘锦高
作者单位: 华东师范大学信息科学与技术学院极化材料与器件教育部重点实验室
摘 要: 使用自行研制的分子光谱成像系统获取了白血病人血液涂片的分子光谱图像数据,用于血细胞的自动识别和分析。针对分子光谱数据具有波段数多、数据量大的特点,提出一种具有自适应性的遗传算法(genetic algorithm, GA)用于血细胞分子光谱图像的波段选择。实验结果表明:这种自适应遗传算法(adaptive genetic algorithm ,AGA)鲁棒性强、搜索效率高,比简单遗传算法(simple genetic algoristhm, SGA)更有利于血细胞图像的解译。
关 键 词: 模式识别;遗传算法;波段选择;高光谱成像;生物医学图像
Title: Band selection for molecular hyperspectral data of blood cells based on adaptive genetic algorithms
Author: DAI Chunni, LI Qingli, LIU Jingao
Organization: Key Laboratory of Polor Materials and Devices, Ministry of Education, School of Information Science and Technology, East China Normal University
Abstract: A molecular hyperspectral imaging device was developed to study the leukemic blood cells of human. As the molecular hyperspectral images have hundreds of wavelength bands and large amount of data, it is very important to perform the dimension reduction process before intelligent recognition and analysis. In this paper, an adaptive genetic algorithm is applied for band selection form the molecular hyperspectral images of blood cells. Experiment results show that the new algorithm is prior to the simple genetic algorithms with the convenience to better understanding and diagnosis.
Key words: pattern recognition;genetic algorithm; band selection; hyperspectral imaging; biomedical images
发表期数: 2009年9月第18期
引用格式: 戴春妮,李庆利,刘锦高. 基于自适应遗传算法的血细胞分子光谱图像波段选择[J]. 中国科技论文在线精品论文,2009,2(18):1943-1948.
 
0 评论数 0
暂无评论
友情链接