基于贝叶斯分类器的图像分类  

Image Classification Based on Bayesian Classifier

作　　者：董立岩[1] 苑森淼[1] 刘光远[2] 贾书洪[3] 

机构地区：[1]吉林大学计算机科学与技术学院,长春130012 [2]吉林大学通信工程学院,长春130022 [3]空军航空大学基础部,长春130022

出　　处：《吉林大学学报（理学版）》2007年第2期249-253,共5页Journal of Jilin University:Science Edition

基　　金：国家自然科学基金(批准号:60275026)

摘　　要：通过对朴素贝叶斯分类器的讨论,提出将贝叶斯方法应用于医学图像分割后的图像分类思想.给出一种基于朴素贝叶斯分类器的图像分类方法,对从尿沉渣图像中识别出的微粒进行正确分割及特征提取与选择,并利用朴素贝叶斯分类器进行分类.实验结果表明,所提出的方法用于解决图像分类有效.Naive Bayesian classifier was discussed. Naive Bayesian approach was used in image classification. The new method improves the accuracy of image recognition. An image classification method based on naive Bayesian classifier was introduced. After distinguishing the visible compositions, feature extraction and feature selection were processed. In the end, naive Bayesian classifier was used in classifying of object entities. Experiment result shows that the new method is suitable for image classification.

关 键 词：贝叶斯分类器 图像分类 特征提取 特征选择 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]