pcl 如何逐步匹配多幅点云

中文教程 http://www.pclcn.org/study/shownews.php?lang=cn&id=78
http://www.pclcn.org/study/shownews.php?lang=cn&id=79
英文原教程
http://pointclouds.org/documentation/tutorials/pairwise_incremental_registration.php#pairwise-incremental-registration
ICP算法的使用，以便逐步地对一系列点云进行两两匹配。它的思想是对所有的点云进行变换，使得都与第一个点云在统一坐标系中。在每个连贯的有重叠的点云之间找到最佳的变换，并累积这些变换到全部的点云。能够进行ICP算法的点云需要粗略的预匹配（如：在一个机器人的量距内或在地图框架内），并且一个点云与另一个点云需要有重叠部分。
code

 /* \author Radu Bogdan Rusu
* adaptation Raphael Favier*/
#include <boost/make_shared.hpp>
#include <pcl/point_types.h>
#include <pcl/point_cloud.h>
#include <pcl/point_representation.h>
#include <pcl/io/pcd_io.h>
#include <pcl/filters/voxel_grid.h>
#include <pcl/filters/filter.h>
#include <pcl/features/normal_3d.h>
#include <pcl/registration/icp.h>
#include <pcl/registration/icp_nl.h>
#include <pcl/registration/transforms.h>
#include <pcl/visualization/pcl_visualizer.h>
using pcl::visualization::PointCloudColorHandlerGenericField;
using pcl::visualization::PointCloudColorHandlerCustom;
//简单类型定义
typedef pcl::PointXYZ PointT;
typedef pcl::PointCloud<PointT> PointCloud;
typedef pcl::PointNormal PointNormalT;
typedef pcl::PointCloud<PointNormalT> PointCloudWithNormals;
//这是一个辅助教程，因此我们可以负担全局变量
//创建可视化工具
pcl::visualization::PCLVisualizer *p;
//定义左右视点
int vp_1, vp_2;
//处理点云的方便的结构定义
struct PCD
{
    PointCloud::Ptr cloud;
    std::string f_name;
    PCD() : cloud(new PointCloud) {};
};
struct PCDComparator
{
    bool operator () (const PCD& p1, const PCD& p2)
    {
        return (p1.f_name < p2.f_name);
    }
};
//以< x, y, z, curvature >形式定义一个新的点
class MyPointRepresentation : public pcl::PointRepresentation <PointNormalT>
{
    using pcl::PointRepresentation<PointNormalT>::nr_dimensions_;
public:
    MyPointRepresentation()
    {
        //定义尺寸值
        nr_dimensions_ = 4;
    }
    //覆盖copyToFloatArray方法来定义我们的特征矢量
    virtual void copyToFloatArray(const PointNormalT &p, float * out) const
    {
        // < x, y, z, curvature >
        out[0] = p.x;
        out[1] = p.y;
        out[2] = p.z;
        out[3] = p.curvature;
    }
};
////////////////////////////////////////////////////////////////////////////////
/** 在可视化窗口的第一视点显示源点云和目标点云
*
*/
void showCloudsLeft(const PointCloud::Ptr cloud_target, const PointCloud::Ptr cloud_source)
{
    p->removePointCloud("vp1_target");
    p->removePointCloud("vp1_source");
    PointCloudColorHandlerCustom<PointT> tgt_h(cloud_target, 0, 255, 0);
    PointCloudColorHandlerCustom<PointT> src_h(cloud_source, 255, 0, 0);
    p->addPointCloud(cloud_target, tgt_h, "vp1_target", vp_1);
    p->addPointCloud(cloud_source, src_h, "vp1_source", vp_1);
    PCL_INFO("Press q to begin the registration.\n");
    p->spin();
}
////////////////////////////////////////////////////////////////////////////////
/**在可视化窗口的第二视点显示源点云和目标点云
*
*/
void showCloudsRight(const PointCloudWithNormals::Ptr cloud_target, const PointCloudWithNormals::Ptr cloud_source)
{
    p->removePointCloud("source");
    p->removePointCloud("target");
    PointCloudColorHandlerGenericField<PointNormalT> tgt_color_handler(cloud_target, "curvature");
    if (!tgt_color_handler.isCapable())
        PCL_WARN("Cannot create curvature color handler!");
    PointCloudColorHandlerGenericField<PointNormalT> src_color_handler(cloud_source, "curvature");
    if (!src_color_handler.isCapable())
        PCL_WARN("Cannot create curvature color handler!");
    p->addPointCloud(cloud_target, tgt_color_handler, "target", vp_2);
    p->addPointCloud(cloud_source, src_color_handler, "source", vp_2);
    p->spinOnce();
}
////////////////////////////////////////////////////////////////////////////////
/**加载一组我们想要匹配在一起的PCD文件
* 参数argc是参数的数量 (pass from main ())
*参数 argv 实际的命令行参数 (pass from main ())
*参数models点云数据集的合成矢量
*/
void loadData(int argc, char **argv, std::vector<PCD, Eigen::aligned_allocator<PCD> > &models)
{
    std::string extension(".pcd");
    //假定第一个参数是实际测试模型
    for (int i = 1; i < argc; i++)
    {
        std::string fname = std::string(argv[i]);
        // 至少需要5个字符长（因为.plot就有 5个字符）
        if (fname.size() <= extension.size())
            continue;
        std::transform(fname.begin(), fname.end(), fname.begin(), (int(*)(int))tolower);
        //检查参数是一个pcd文件
        if (fname.compare(fname.size() - extension.size(), extension.size(), extension) == 0)
        {
            //加载点云并保存在总体的模型列表中
            PCD m;
            m.f_name = argv[i];
            pcl::io::loadPCDFile(argv[i], *m.cloud);
            //从点云中移除NAN点
            std::vector<int> indices;
            pcl::removeNaNFromPointCloud(*m.cloud, *m.cloud, indices);
            models.push_back(m);
        }
    }
}
////////////////////////////////////////////////////////////////////////////////
/**匹配一对点云数据集并且返还结果
*参数 cloud_src 是源点云
*参数 cloud_src 是目标点云
*参数output输出的配准结果的源点云
*参数final_transform是在来源和目标之间的转换
*/
void pairAlign(const PointCloud::Ptr cloud_src, const PointCloud::Ptr cloud_tgt, PointCloud::Ptr output, Eigen::Matrix4f &final_transform, bool downsample = false)
{
    //
    //为了一致性和高速的下采样
    //注意：为了大数据集需要允许这项
    PointCloud::Ptr src(new PointCloud);
    PointCloud::Ptr tgt(new PointCloud);
    pcl::VoxelGrid<PointT> grid;
    if (downsample)
    {
        grid.setLeafSize(0.05, 0.05, 0.05);
        grid.setInputCloud(cloud_src);
        grid.filter(*src);
        grid.setInputCloud(cloud_tgt);
        grid.filter(*tgt);
    }
    else
    {
        src = cloud_src;
        tgt = cloud_tgt;
    }
    //计算曲面法线和曲率
    PointCloudWithNormals::Ptr points_with_normals_src(new PointCloudWithNormals);
    PointCloudWithNormals::Ptr points_with_normals_tgt(new PointCloudWithNormals);
    pcl::NormalEstimation<PointT, PointNormalT> norm_est;
    pcl::search::KdTree<pcl::PointXYZ>::Ptr tree(new pcl::search::KdTree<pcl::PointXYZ>());
    norm_est.setSearchMethod(tree);
    norm_est.setKSearch(30);
    norm_est.setInputCloud(src);
    norm_est.compute(*points_with_normals_src);
    pcl::copyPointCloud(*src, *points_with_normals_src);
    norm_est.setInputCloud(tgt);
    norm_est.compute(*points_with_normals_tgt);
    pcl::copyPointCloud(*tgt, *points_with_normals_tgt);
    //
    //举例说明我们自定义点的表示（以上定义）
    MyPointRepresentation point_representation;
    //调整'curvature'尺寸权重以便使它和x, y, z平衡
    float alpha[4] = { 1.0, 1.0, 1.0, 1.0 };
    point_representation.setRescaleValues(alpha);
    //
    // 配准
    pcl::IterativeClosestPointNonLinear<PointNormalT, PointNormalT> reg;
    reg.setTransformationEpsilon(1e-6);
    //将两个对应关系之间的(src<->tgt)最大距离设置为10厘米
    //注意：根据你的数据集大小来调整
    reg.setMaxCorrespondenceDistance(0.1);
    //设置点表示
    reg.setPointRepresentation(boost::make_shared<const MyPointRepresentation>(point_representation));
    reg.setInputCloud(points_with_normals_src);
    reg.setInputTarget(points_with_normals_tgt);
    //
    //在一个循环中运行相同的最优化并且使结果可视化
    Eigen::Matrix4f Ti = Eigen::Matrix4f::Identity(), prev, targetToSource;
    PointCloudWithNormals::Ptr reg_result = points_with_normals_src;
    reg.setMaximumIterations(2);
    for (int i = 0; i < 30; ++i)
    {
        PCL_INFO("Iteration Nr. %d.\n", i);
        //为了可视化的目的保存点云
        points_with_normals_src = reg_result;
        //估计
        reg.setInputCloud(points_with_normals_src);
        reg.align(*reg_result);
        //在每一个迭代之间累积转换
        Ti = reg.getFinalTransformation() * Ti;
        //如果这次转换和之前转换之间的差异小于阈值
        //则通过减小最大对应距离来改善程序
        if (fabs((reg.getLastIncrementalTransformation() - prev).sum()) < reg.getTransformationEpsilon())
            reg.setMaxCorrespondenceDistance(reg.getMaxCorrespondenceDistance() - 0.001);
        prev = reg.getLastIncrementalTransformation();
        //可视化当前状态
        showCloudsRight(points_with_normals_tgt, points_with_normals_src);
    }
    //
    // 得到目标点云到源点云的变换
    targetToSource = Ti.inverse();
    //
    //把目标点云转换回源框架
    pcl::transformPointCloud(*cloud_tgt, *output, targetToSource);
    p->removePointCloud("source");
    p->removePointCloud("target");
    PointCloudColorHandlerCustom<PointT> cloud_tgt_h(output, 0, 255, 0);
    PointCloudColorHandlerCustom<PointT> cloud_src_h(cloud_src, 255, 0, 0);
    p->addPointCloud(output, cloud_tgt_h, "target", vp_2);
    p->addPointCloud(cloud_src, cloud_src_h, "source", vp_2);
    PCL_INFO("Press q to continue the registration.\n");
    p->spin();
    p->removePointCloud("source");
    p->removePointCloud("target");
    //添加源点云到转换目标
    *output += *cloud_src;
    final_transform = targetToSource;
}
/* ---[ */
int main(int argc, char** argv)
{
    // 加载数据
    std::vector<PCD, Eigen::aligned_allocator<PCD> > data;
    loadData(argc, argv, data);
    //检查用户输入
    if (data.empty())
    {
        PCL_ERROR("Syntax is: %s <source.pcd> <target.pcd> [*]", argv[0]);
        PCL_ERROR("[*] - multiple files can be added. The registration results of (i, i+1) will be registered against (i+2), etc");
        PCL_INFO("Example: %s `rospack find pcl`/test/bun0.pcd `rospack find pcl`/test/bun4.pcd", argv[0]);
        return (-1);
    }
    PCL_INFO("Loaded %d datasets.", (int)data.size());
    //创建一个PCL可视化对象
    p = new pcl::visualization::PCLVisualizer(argc, argv, "Pairwise Incremental Registration example");
    p->createViewPort(0.0, 0, 0.5, 1.0, vp_1);
    p->createViewPort(0.5, 0, 1.0, 1.0, vp_2);
    PointCloud::Ptr result(new PointCloud), source, target;
    Eigen::Matrix4f GlobalTransform = Eigen::Matrix4f::Identity(), pairTransform;
    for (size_t i = 1; i < data.size(); ++i)
    {
        source = data[i - 1].cloud;
        target = data[i].cloud;
        //添加可视化数据
        showCloudsLeft(source, target);
        PointCloud::Ptr temp(new PointCloud);
        PCL_INFO("Aligning %s (%d) with %s (%d).\n", data[i - 1].f_name.c_str(), source->points.size(), data[i].f_name.c_str(), target->points.size());
        pairAlign(source, target, temp, pairTransform, true);
        //把当前的两两配对转换到全局变换
        pcl::transformPointCloud(*temp, *result, GlobalTransform);
        //update the global transform更新全局变换
        GlobalTransform = pairTransform * GlobalTransform;
        //保存配准对，转换到第一个点云框架中
        std::stringstream ss;
        ss << i << ".pcd";
        pcl::io::savePCDFile(ss.str(), *result, true);
    }
}
/* ]--- */