基于SVM目标识别的图像特征选取研究  

Image Feature Selection for Military Target Recognition Based on SVM

作　　者：王一波 宋小杉 WANG Yibo;SONG Xiaoshan(Liuzhou Institute of Technology,Liuzhou 545000,China;Rocket Force University of Engineering,Xi'an 710000,China)

机构地区：[1]柳州工学院,广西柳州545000 [2]火箭军工程大学,西安710000

出　　处：《电光与控制》2021年第12期36-39,51,共5页Electronics Optics & Control

基　　金：国家自然科学基金(54105510);广西科技基地和人才专项(2020AC19115);广西高校中青年教师科研基础能力提升项目(2020KY60013)。

摘　　要：基于SVM的作战目标识别具有速度快、鲁棒性好的特点,其中图像特征选取是准确识别的关键。针对野战环境下目标图像所受干扰因素复杂的问题,提出复合式特征选取方法。对获取的可见光和红外图像,首先通过图像评价筛选有判别价值的目标图像,基于Soft方法进行图像分割,接着提取不变矩特征,最后进行特征融合,为SVM提供可靠的特征向量。SVM-based military target recognition has the characteristics of fast speed and good robustnessand accurate recognition relies on image feature selection.In field environmentthe target's image is affected by complex interference factors.In view of the problema compound feature selection method is proposed.The visible and infrared images obtained are evaluatedin which the discriminative ones are selected and segmented based on Soft method.Thenmoment-invariant features are extracted and feature fusion is made to provide reliable feature vectors for SVM.

关 键 词：目标识别 作战目标 支持向量机 特征选取 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]