pc_point= pc.Location'; %得到点云数据kdtree =vl_kdtreebuild(pc_point); %使用vlfeat建立kdtree dissum=0;fori=1:length(pc_point) p_cur=pc_point(:,i); [index, distance]= vl_kdtreequery(kdtree, pc_point, p_cur,'NumNeighbors',2); %寻找当前点最近的非自身点 dissum= dissum + sq...
1clear all;2close all;3clc;45pc = pcread('rabbit.pcd');6pc = pcdownsample(pc,'random',0.1); %降低一下数据量7pc_point = pc.Location'; %得到点云数据8kdtree = vl_kdtreebuild(pc_point); %使用vlfeat建立kdtree910dissum =0;11fori=1:length(pc_point)12p_cur =pc_point(:,i);13[...
1clear all;2close all;3clc;45pc = pcread('rabbit.pcd');6pc = pcdownsample(pc,'random',0.1); %降低一下数据量7pc_point = pc.Location'; %得到点云数据8kdtree = vl_kdtreebuild(pc_point); %使用vlfeat建立kdtree910dissum =0;11fori=1:length(pc_point)12p_cur =pc_point(:,i);13[...
kdtree = vl_kdtreebuild(X); % 计算样本X(:,i)的k个近邻,并构造邻接矩阵A for i = 1 : band [index, ~] = vl_kdtreequery(kdtree, X, X(:,i), 'NumNeighbors', 6) ; % 排除样本本身 index = index(index ~= i); % 构造邻接矩阵A,此处可以体会邻接矩阵为整个高光谱图的邻接矩阵 A(i...
pc_point= pc.Location'; %得到点云数据kdtree =vl_kdtreebuild(pc_point); %使用vlfeat建立kdtree normE=[];fori=1:length(pc_point) p_cur=pc_point(:,i); [index, distance]= vl_kdtreequery(kdtree, pc_point, p_cur,'NumNeighbors',10); %寻找当前点最近的10个点 ...
clear all; close all; clc; warning off; pc = pcread('rabbit.pcd'); pc=pcdownsample(pc,'random',0.3); %0.3倍降采样 pcshow(pc); pc_point = pc.Location'; %得到点云数据 kdtree = vl_kdtreebuild(pc_point); %使用vlfeat建立kdtree normE=[]; for i=1:length(pc_point) p_cur = pc...
2. 在matlab中输入vl_version,可以得到vlfeat的版本号。 有这些东西: • TheVLFeatlibrary -SIFTexample (vl_sift) • Caltech-101running example • Visual descriptors -PHOWfeature (fast dense SIFT, vl_phow) - Vector Quantization (Elkan, vl_kmeans, vl_kdtreebuild, ...