一种基于特征融合的车辆图像分类方法  

A Method of Vehicle Image Classification Based on Multi-feature Fusion

作　　者：邓方舟 何南[1] Deng Fangzhou;He Nan(Center for Network and Educational Technology,Guilin Normal College,Guilin,Guangxi 541199,China)

机构地区：[1]桂林师范高等专科学校网络与教育技术中心,广西桂林541199

出　　处：《桂林师范高等专科学校学报》2020年第3期132-135,共4页Journal of Guilin Normal College

基　　金：2018年桂林师范高等专科学校科研项目“基于图像的分类算法和可逆信息隐藏算法研究”(项目编号:KYB201816)。

摘　　要：由于利用单一图像特征描述图像的属性不完全,比较片面,文章在研究将几种不同特征融合进行图像分类的基础上,提出了一种基于Gist特征和颜色矩特征的图像分类方法,并将其应用到车辆图像上。首先对图像分别提取Gist特征、颜色矩特征和Hu不变矩特征,然后将三种特征任意融合,形成特征向量,用libSVM分类器进行训练和分类,确定分类效果最好的融合特征:Gist特征和颜色矩特征。研究结果表明:基于特征融合的车辆图像分类方法效果比单一特征分类效果有所提高,提高了分类的准确率。This paper proposed a multi-feature image classification method based on GIST feature and color moments, and applied it to vehicle images. Initially, image GIST features, color moment features and Hu invariant moment features are extracted respectively. The three features are connected by a serial method in any combination ways. The fusion of GIST features and color moment features has the advantage over other combining methods. The experimental results show that the image classification based on the proposed method is improved compared with other methods.

关 键 词：颜色矩 GIST HU不变矩 特征融合 图像分类 SVM 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]