海量小差异图像群中有效分类方法研究与仿真
Huge Amounts of Small Differences in Effective Classification Method in Image Group Research and Simulation
-
摘要: 对海量小差异图像群进行有效分类,能够极大地满足用户对于图像搜索及管理需求.传统的海量小差异图像分类的方法需要限定大量的分类约束条件,存在分类效率低,无效分类多的弊端.为此,提出基于模糊贴近度改进的海量小差异图像分类方法.利用SIFT算法对采集的海量小差异图像群进行特征提取分析,为进行图像进一步分类提供数据支持.把模糊贴近度的概念引入到分类算法中,对海量小差异图像群进行分类模型构建,求取最优解,完成对海量小差异图像群的最优分类.实验结果表明,运用改进分类算法对海量小差异图像进行分类,能够提高图像分类的效率和分类的准确率.Abstract: To effective mass of difference image classification,can greatly satisfy the user demand for image search and management.Traditional mass small differences in classification of image classification method to limit a lot of constraint conditions,the existing classification efficiency is low,the invalid classification more disadvantages.Therefore,based on fuzzy joint degree improved mass small difference image classification method.Using the SIFT algorithm for feature extraction with the mass of difference image of analysis,provide data support for the further image classification.The concept of the fuzzy degree of joint is introduced into the classification algorithm,the mass of difference image classification model build,calculate the optimal solution,to the optimal mass of difference image classification.The experimental results show that the improved classification algorithm is used to mass small difference image classification,to improve the efficiency of image classification and classification accuracy.
下载:
