基于扩展Shearlet变换、Krawtchouk矩和SVM的储粮害虫分类  

A Classification Method of Stored-Grain Pests Based on the Extended Shearlet Transform,Krawtchouk Moment and SVM

作　　者：吴一全[1,2] 王凯[1] 陶飞翔 

机构地区：[1]南京航空航天大学电子信息工程学院,南京210016 [2]南京财经大学江苏省粮油品质控制及深加工技术重点实验室,南京210046

出　　处：《中国粮油学报》2015年第11期103-109,共7页Journal of the Chinese Cereals and Oils Association

基　　金：江苏省粮油品质控制及深加工技术重点实验室开放基金(LYPK201304);江苏高校优势学科建设工程资助(20111008)

摘　　要：为了进一步提高储粮害虫的识别精度,以便更有效地防治储粮害虫,提出了一种基于纹理和形状综合特征及全局混沌蜂群优化支持向量机(SVM)的储粮害虫分类方法。首先对储粮害虫图像进行扩展Shearlet变换,利用变换系数得到能量分布均值,加权后的能量分布均值构成纹理特征向量,用Krwtchouk矩不变量描述储粮害虫的形状特征;然后将纹理特征向量和形状特征向量分别归一化,两者结合构成储粮害虫的综合特征向量;最后用全局混沌蜂群算法优化SVM的核参数与惩罚因子,并应用参数优化的SVM进行分类。结果表明:与基于Gabor小波和支持向量机方法、基于Krawtchouk不变矩和支持向量机方法相比,本方法提取的储粮害虫特征信息更加完整,识别率更高。To further improve the recognition accuracy of stored - grain pests and control stored - grain pests in a more effective way, a classification method of stored - grain pests has been proposed based on the synthetic features combining texture feature with shape feature and Support Vector Machine （SVM） to be optimized by global chaotic bee colony algorithm. First, an image with stored -grain pests was decomposed by the extended shearlet transform; means of energy distribution were calculated by transform coefficients; texture feature vector was produced with the weighted energy distribution. The shape feature vector of stored - grain pests was represented by Krawtchouk moment invariants. Second, the normalized texture feature vector and the normalized shape feature vector were combined to form the synthetic feature vector of stored -grain pests. Finally, the kernel parameter and the penalty factor of sup- port vector machine were optimized with global chaotic bee colony algorithm; the optimized support vector machine was applied for classification of stored - grain pests. The experimental results showed that compared with the method based on Gabor wavelet and SVM and the method based on Krawtchouk moment invariants and SVM, the proposed method extracted more complete characteristic information of stored - grain pests with higher recognition rate.

关 键 词：储粮害虫分类 纹理特征 形状特征 扩展Shearlet变换 Krawtchouk矩不变量 支持向量机 全局混沌蜂群算法 

分 类 号：S24[农业科学—农业电气化与自动化] TP391.41[农业科学—农业工程]