局部图像描述符最新研究进展  

Recent advances in local image descriptor

作　　者：许允喜[1] 陈方[1] 

机构地区：[1]湖州师范学院信息工程学院,湖州313000

出　　处：《中国图象图形学报》2015年第9期1133-1150,共18页Journal of Image and Graphics

基　　金：国家自然科学基金项目(61370173);湖州市重点科技创新团队(2012KC04)

摘　　要：目的局部图像描述符广泛应用于许多图像理解和计算机视觉应用领域,如图像分类、目标识别、图像检索、机器人导航、纹理分类等。SIFT算法的提出标志着现代局部图像描述符研究的开始。主要对最近发展的现代局部图像描述符进行了综述。方法首先,介绍了4大类局部图像描述符:局部特征空间分布描述符、局部特征空间关联描述符、基于机器学习的局部描述符、扩展局部描述符(局部颜色描述符、局部RGB-D描述符、局部空时描述符)。对局部图像描述符进行了分析和分类,并总结了局部图像描述符的不变性、计算复杂度、应用领域、评价方法和评价数据集。最后,展望了局部图像描述符的未来研究方向。结果近年来局部图像描述符研究取得了很大进展,提出了很多优秀的描述符,在辨别性、鲁棒性和实时性方面有了很大提高,应用领域不断拓展。结论局部图像描述符应用广泛,是计算机视觉领域的重要基础研究。而目前,局部图像描述符还存在许多问题,还需进一步的深入研究。Objective Local image descriptors are widely utilized in many image understanding and computer vision applica- tions, such as image classification, object recognition, image retrieval, robot navigation, and texture classification. The de- velopment of the SIFT algorithm highlighted the beginning of modem local image descriptor research. Recently developed modem local image descriptors are surveyed in this study. Method Four types of local image descriptors, namely, spatial distribution descriptors of local features, spatial correlation descriptors of local features, local descriptors based on machine learning, and extended local descriptors （local color descriptors, local RGB-D descriptors, and local space-time descrip- tors）, are presented. The local image descriptors are then analyzed and categorized. The performance of the local image descriptors is investigatedin terms of invariance, computation complexity, application field, evaluation methods, and evalu- ation datasets. Finally, this paper concludes with a discussion of directions for future research of the local image descrip- tors. Result In recent years, the research of local image descriptors has made great progress. Many excellent descriptors are proposed l and their performances have greatly improved in the distinctiveness, robustness and real-time, and their ap- plication fields are continually expanded. Conclusion Local image descriptors are widely utilized as an important and fun- damental research field of computer vision. However, many problems in the use of these descriptors persist. This condition indicates that further research on local image descriptors is required.

关 键 词：局部图像描述符 局部不变特征 图像匹配 图像分类 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]