针对传统的视觉词袋(bag of visual words,BoVW)模型忽略了视觉单词的空间位置信息的问题,文章提出一种基于视觉单词共生矩阵的图像分类方法。首先对整幅图像进行空间金字塔分解,得到一系列图像块;然后针对每一图像块中的SIFT点,在其空...针对传统的视觉词袋(bag of visual words,BoVW)模型忽略了视觉单词的空间位置信息的问题,文章提出一种基于视觉单词共生矩阵的图像分类方法。首先对整幅图像进行空间金字塔分解,得到一系列图像块;然后针对每一图像块中的SIFT点,在其空间邻域范围内构建视觉单词共生矩阵(visual words co-occurrence matrix,VWCM)单元,并得到该图像块对应的视觉单词共生矩阵;最后设计出一种新的空间金字塔共生矩阵核(spatial pyramid co-occurrence matrix kernel,SPCMK),并将其用于图像分类。该方法能够有效地刻画视觉单词的绝对和相对位置信息,极大地增强了图像表达的完整度与准确度。实验结果表明,文章方法确实能够大幅度提高图像分类的准确率。展开更多
在结构化场景的轨道交通中,车载视频观测因相机平移运动而呈现出图像内容以某点为中心向四周扩散的现象,该点被称为FOE(Focus of Expansion)。当前计算FOE的算法对噪声敏感且计算量大,不能准确地计算铁路场景中的FOE。鉴于此,文中提出...在结构化场景的轨道交通中,车载视频观测因相机平移运动而呈现出图像内容以某点为中心向四周扩散的现象,该点被称为FOE(Focus of Expansion)。当前计算FOE的算法对噪声敏感且计算量大,不能准确地计算铁路场景中的FOE。鉴于此,文中提出一种铁路视频序列的FOE估计方法。该方法首先利用金字塔光流法对检测的Harris角点进行跟踪和粗匹配,并在此基础上利用RANSAC算法进行精确的匹配,求得基础矩阵,然后提取图像中的极线束并计算FOE。实验结果表明,所提算法比Hough直线求得的FOE误差小,适于实时应用。展开更多
The work presented previously by the authors(Cai and Liou,1982)has been extended in this paper.By making use of our improved model the calculations on scattering phase matrices of hexagonal prismice crystals(HPIC)have...The work presented previously by the authors(Cai and Liou,1982)has been extended in this paper.By making use of our improved model the calculations on scattering phase matrices of hexagonal prismice crystals(HPIC)have been conducted for monodisperse and polydisperse systems.Compared with themodel of Cai and Liou,the required computational quantity is decreased by about two orders of magni-tude and the errors of results are less for the new model.Meanwhile,the scattering phase matrices of triangularpyramid ice crystals(TPIC)are also computed in the paper,and the comparison between the scatterings of thetwo forms of ice crystals is performed.展开更多
文摘在结构化场景的轨道交通中,车载视频观测因相机平移运动而呈现出图像内容以某点为中心向四周扩散的现象,该点被称为FOE(Focus of Expansion)。当前计算FOE的算法对噪声敏感且计算量大,不能准确地计算铁路场景中的FOE。鉴于此,文中提出一种铁路视频序列的FOE估计方法。该方法首先利用金字塔光流法对检测的Harris角点进行跟踪和粗匹配,并在此基础上利用RANSAC算法进行精确的匹配,求得基础矩阵,然后提取图像中的极线束并计算FOE。实验结果表明,所提算法比Hough直线求得的FOE误差小,适于实时应用。
文摘The work presented previously by the authors(Cai and Liou,1982)has been extended in this paper.By making use of our improved model the calculations on scattering phase matrices of hexagonal prismice crystals(HPIC)have been conducted for monodisperse and polydisperse systems.Compared with themodel of Cai and Liou,the required computational quantity is decreased by about two orders of magni-tude and the errors of results are less for the new model.Meanwhile,the scattering phase matrices of triangularpyramid ice crystals(TPIC)are also computed in the paper,and the comparison between the scatterings of thetwo forms of ice crystals is performed.