基于伪Zernike矩的羽毛缺陷几何特征提取及识别  

Geometric Feature Extraction and Classification Examination of Feather Defects Based on Pseudo-Zernike Moments

作　　者：明俊峰[1] 汪仁煌[1] 岳洪伟[2] 唐苏湘[1] 

机构地区：[1]广东工业大学自动化学院,广州510090 [2]五邑大学信息工程学院,广东江门529020

出　　处：《四川兵工学报》2014年第9期69-72,91,共5页Journal of Sichuan Ordnance

基　　金：广东省自然科学基金重点项目(S2012020011071);广东省战略性新兴产业核心技术攻关项目(2012A010701004)

摘　　要：针对拍照角度影响羽毛缺陷识别精度的问题,将不变矩的思想引入到羽毛缺陷识别领域。提出利用伪Zernike矩描述羽毛缺陷的几何特征,用支持向量机(SVM)进行分类识别的方法。模拟实验表明算法能够较好地容忍图像平移、旋转和尺度变化,具有识别速度快、识别精度高的特点,并且算法对噪声、光线变化适应性好。Aimed at the recognition accuracy problem of feather defects caused by the impacts of camera angle ,the idea of invariant moment is introduced into feather defect recognition.To identify the feather de-fects,the method based on support vector machine （SVM）,by extracting feature vector with Pseudo-Zernike moment,was proposed.Simulation experiments show that the algorithm can better tolerate the im-age translation,rotation and scaling transformation with characteristics of fast recognition speed and high recognition accuracy,and it is also good adaptability for noise and light changes.

关 键 词：羽毛缺陷 几何特征 伪ZERNIKE矩 SVM分类器 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]