摘要
针对如何能更有效地检索出目标图像,提出一种基于形状和遗传算法的图像检索方法。利用遗传算法具有群体选优的特性,对图像的统计特征进行特征选取,获取最佳匹配特征以便检索出最能满足用户需求的结果。以Corel图像库中的部分图像为测试集,在VC++6.0环境下进行实验。实验结果表明,利用遗传算法进行图像检索能够有效地进行选取特征,大幅度地高了检索结果的查准率和查全率,并缩短了检索时间。与基于粒子群优化算法的检索结果进行比较,发现利用遗传算法的检索结果优于后者。
Aims at efficiently retrieving the target images, an image retrieval method based on Shape and Genetic Algorithm is proposed in this paper. Statistic features are selected by Genetic Algorithm that owns the characteristics of selecting the optimal features, and get the features of the best matches to meet the requirement of the users, which is applied in Corel image database and the VC ＋ ＋ 6.0. The experiment results show the Genetic Algorithm can efficiently select features, greatly improve the precision and recall , and shorten the retrieval time. At last, it is made comparison with the Particle Swarm Optimization Algorithm, which appears the retrieval results of the Genetic Algorithm are better than the Particle Swarm Optimization Algorithm.
出处
《计算技术与自动化》
2008年第4期98-101,共4页
Computing Technology and Automation
关键词
图像检索
遗传算法
粒子群优化
查准率
查全率
image retrieval
genetic algorithm
particle swarm optimization
precision
recall
