搜索资源列表
distance
- 关于计算机坐标和地球坐标的变换,可以用于图像处理中标线及划分区域。-coordinates on the computer and transform the coordinates of the Earth that can be used in image processing and successful line separating regions.
distance transform
- 目前效率最好的距离变换算法,距离变换广泛应用于图像处理,比如可以用距离变换准确找到图像中物体的骨架或者中心线。(用winrar解压)-efficiency of the current best distance algorithm distance transform widely used image processing, For example, can be used to find accurate distance transform image objects in the fr
jiyuHausdorffdetuxiangpeizhun.
- 针对图像配准中常出现的RST(旋转-比例-平移)变换,推导出了相应的盒距离变换公式。与传统的基于广 义仿射变换的Hausdorff盒距离变换公式相比,缩小了搜索距离空间。在计算Voronoi表面时,根据Hausdorff距 离的计算需要提出比较滑动窗口的区域Voronoi表面,节省了计算Voronoi表面的时间。并且在利用边缘点计算 Hausdorff距离时,剔除琐碎的边缘,仅使用较长的边缘计算。试验结果表明,这些改进方法较大地提高了基于 Hausdorff距离的图像配准的计算速度。
123
- 一种基于方向信息的鲁棒型Hausdorff距离匹配方法。该方法采用方向信息提取图 像边缘,通过计算边缘匹配率( edge matching rate, EMR)获得候选匹配区域,然后采用修正后的Hausdorff距离构造 相似性测度。实验结果表明,该方法加快了匹配过程,提高了抗噪性能,并能够准确匹配含有遮挡和伪边缘点的图 像,从而解决了基于传统Hausdorff距离匹配方法因噪声点、伪边缘点和出格点而造成的误匹配问题。-Based on the direction of inform
An.Image.Mosaic.Algorithm.Based.on.Matching.Shape.
- 提 出了一种基于形状模板 匹配的图像 自动拼接 方法。提取 图像 的角点作 为特 征点,利用归一化梯度模板对 其进行预 匹配.然后 利用形状模板在 四个方 向对模板 内图像的边缘 点与模板边界的最短距 离进行统计 ,获取模板 图像的 结构特征向量以实现对特征点的精确匹配。实验结果表明该算法具有较好的实用价值。 -A shape-based template matching method of image auto-splicing. Extract image feature po
Image
- 遥感图像的打开处理程序,能够打开单波段,多波段图像,几何图像变换,线性拉伸变换,平滑 处理包括并行和串行,锐化处理包括梯度锐化、Roberts锐化、laplace锐化、sobel锐化等,还 有用绝对距离和马氏距离算法进行的监督分类算法等,包括了RAW格式数据资源-The opening of remote sensing image processing, to open the single-band, multi-band images, geometric image
a-level-set-inariable-distance
- 李春明提出的无需水平集初始化的思想是图像分割的一个重大突破,重庆大学的何传江教授对该算法进行了改进,里面是我的实现算法-Li Chunming s no need to initialize the idea of the level set image segmentation is a major breakthrough, Chongqing University, Professor He Chuanjiang to improve the algorithm
DisTrans
- 快速算法,可以快速计算一个二进制图片的Euclidean距离。-Fast algorithm can quickly calculate a binary image of the Euclidean distance.
Object_Removal_by_Exemplar-Based_Inpainting
- 基于样例的图像修补方法,可以用于移除较大的目标,修复较大的区域。VC++环境,Qiushuang Zhang设计,参见A.Criminisi等人的论文:Object Removal by Exemplar-Based Inpainting.-Based on the sample image repair method can be used to remove larger goal, to repair the larger region. VC++ Environment, Qiushua
200707171152015173
- 图像检索中颜色的特征提取及匹配算法,以家权欧几里得距离,中心距得加权距离,直方图交集算法等。-Image Retrieval color feature extraction and matching algorithm to the right home Euclid distance, center distance of a weighted distance, histogram intersection algorithm.
cbir
- 用的是局部颜色特征,再说细点是用里面的区域颜色直方图的方法。把图像归一化到256X256,把图像分成4X4块,计算16个区域的颜色直方图、、、 最后计算相似度是用欧氏距离.-Using local color feature, repeat fine-point is inside the regional color histogram method. The normalized image to 256X256, the image is divided into 4X4 blocks
Hausdorff.tar
- Hausdorff Distance for Image Recognition
image-retrieval
- 基于颜色进行图像检索的一个简单实现 使用欧氏距离-Color-based image retrieval using the implementation of a simple Euclidean distance
ActivityRecognition
- Activity recognition program using Motion History Image s 7 Hu moments. The program compare the Mahalanobis distance of the Hu moments of the video input with the reference and find out the activities.
DistanceTransform
- 对给定的图像做距离变换,用于图像中目标定位-do distance transform for an image
ToolBox
- matlab图像处理工具相,使用了主成分分析,ANN,SVM等方法。-This toolBox used in the image processing(feature extraction and classification) PCA,LDA,ICA,DCT,RBF,RBE,GRNN,KNN,minimum distance,SVM, and others
templatematching
- 基于hausdorff距离的图像模板匹配算法-Hausdorff distance image based on template matching algorithm
kmeans-image-segmentation
- K-means算法是一种动态聚类方法,这种方法先选择若干样本作为聚类的中心,在按某种聚类准则(通常采用最小距离原则)使各种样本向各个中心积聚,从而得到初始的分类,然后,判断分类的合理性,如果不合理,就修改分类,如此反复的修改聚类的迭代运算,直到合理为止。-K-means algorithm is a dynamic clustering method, this method, select the number of samples as a cluster center in the clu
image-segment
- 基于最大类间最大距离比准则的图像分割,基于最大类间最大距离准则的图像分割,最大熵准则的图像分割,这三种方法的源代码-The maximum distance between two classes based on the maximum ratio criterion for image segmentation, based on the maximum between the maximum distance criterion for image segmentation, maximu
color image segmentation
- Clustering is a way to separate groups of objects. K-means clustering treats each object as having a location in space. It finds partitions such that objects within each cluster are as close to each other as possible, and as far from objects in other