摘要
针对单一特征难以精确地表达复杂图像内容的问题,提出基于颜色通道融合特征的现勘图像分类算法。首先,分别在H,S和V三个颜色通道上提取图像的LBP特征和GIST特征,并利用颜色空间信息进行加权融合;然后,将融合的LBP和GIST特征串联形成新的特征描述向量,并用于训练分类器以实现精确地现勘图像分类。在现勘图像数据库上,大量实验结果显示提出的现勘图像分类算法优于基于单一特征的图像分类正确率。
A crime scene investigation(CSI)classification algorithm based on color channel fusion feature is proposed to solve the problem that it is difficult for a single feature to accurately express the complex image content.The LBP features and GIST features of the image are extracted from the three color channels of H,S and V respectively,and the weight fusion of them is performed with the color space information.The fused LBP and GIST features are connected in series to form a new feature description vector,which is used to train the classifier to realize accurate classification of the CSI image.On the CSI image database,a large number of experimental results show that the classification accuracy of the proposed CSI image classification algorithm is better than that of image classification based on a single feature.
作者
刘颖
张倩楠
王富平
雷研博
公衍超
杨凡超
LIU Ying;ZHANG Qiannan;WANG Fuping;LEI Yanbo;GONG Yanchao;YANG Fanchao(Institute of Image and Information Processing,Xi’an University of Posts and Telecommunications,Xi’an 710121,China;Key Laboratory of Electronic Information Application Technology for Scene Investigation,Ministry of Public Security,Xi’an 710121,China;Key Laboratory of Spectral Imaging Technology,Xi’an Institute of Optics and Precision Mechanics,Chinese Academy of Sciences,Xi’an 710119,China)
出处
《现代电子技术》
北大核心
2020年第4期67-72,共6页
Modern Electronics Technique
基金
国家自然科学基金资助项目(61671733)：国家自然科学基金资助项目(61802305)
西安邮电大学创新基金项目(CXJJLI2018012)
关键词
现勘图像分类
颜色通道
特征提取
特征融合
训练分类器
实验分析
crime scene investigation image classification
color channel
feature extraction
feature fusion
training classifier
experimental analysis
