基于稀疏表示的多特征融合害虫图像识别  

Pest Image Recognition of Multi-feature Fusion Based on Sparse Representation

作　　者：胡永强[1] 宋良图[2] 张洁[2] 谢成军[2] 李瑞[2] 

机构地区：[1]青海省科技信息研究所,西宁810001 [2]中国科学院合肥智能机械研究所,合肥230031

出　　处：《模式识别与人工智能》2014年第11期985-992,共8页Pattern Recognition and Artificial Intelligence

基　　金：国家科技支撑计划项目(No.2012BAK17B15)资助

摘　　要：为提高害虫图像识别的准确率,针对不同害虫具有不同的颜色、形状、纹理的特点,提出一种将颜色、形状、纹理特征与稀疏表示相融合的害虫识别方法.该方法利用已标注的训练样本构造不同特征下的训练样本矩阵,通过求解样本的最优稀疏系数以实现害虫图像识别.由于相同样本通过不同特征训练字典求解的稀疏系数不同,进而识别结果也不同.因此,文中进一步通过设计不同特征下的识别分类器实现多特征的融合.在实验室环境与农田环境下的实验结果表明,相较于其他方法,该方法的害虫识别率获得较大的提高.Aiming at the characteristics of different pest images with different colors, shapes and textures, a pest recognition method based on sparse representation and multi-feature fusion is proposed, which uses a matrix of labeled training samples to construct different dictionaries. The recognition result is achieved by solving optimal sparse coefficients with the corresponding feature dictionary. Furthermore, a novel learning method, which can be improved efficiently by jointly optimizing classifier weights, is presented to effectively fuse multiple features for pest categorization. The experimental results on real datasets show that the proposed method performs well on pest species recognition either in laboratory or in farmland.

关 键 词：害虫识别 特征提取 稀疏表示 多特征融合 ADABOOST 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]