点云处理STEP 1 ：Kinect+PCL 获取原始点云

论文方向是点云处理，最近几个月要参加实习，所以用博客随时记录论文的进程。

这是使用Kinect v2 和 pcl 1.8.0 获取的物体的点云深度图像：

#include <vtkAutoInit.h>
#include <Windows.h>
#include <iostream>
#include <kinect.h>
#include <pcl/io/pcd_io.h> 
#include <pcl/point_types.h> 
#include <pcl/visualization/pcl_visualizer.h>
#include <pcl/filters/radius_outlier_removal.h>  
#include <pcl/filters/conditional_removal.h>
#include <pcl/filters/statistical_outlier_removal.h> 
#include <string.h> 

using namespace std;
typedef pcl::PointXYZ  MyPointDataType;

template<class Interface>
//安全释放指针
inline void SafeRelease(Interface *& pInterfaceToRelease)
{
	if (pInterfaceToRelease != NULL)
	{
		pInterfaceToRelease->Release();
		pInterfaceToRelease = NULL;
	}
}

string num2str(int i)
{
	stringstream ss;
	ss << i;
	return ss.str();
}
//主函数
int main()
{
        //调用Kinect库函数获取Kinect传感器
	IKinectSensor* m_pKinectSensor = nullptr;
	HRESULT hr;
	hr = GetDefaultKinectSensor(&m_pKinectSensor);
	if (FAILED(hr))
	{
		return hr;
	}

	IMultiSourceFrameReader* m_pMultiFrameReader = nullptr;
	if (m_pKinectSensor)
	{
		hr = m_pKinectSensor->Open();
		if (SUCCEEDED(hr))
		{

			hr = m_pKinectSensor->OpenMultiSourceFrameReader(
				FrameSourceTypes::FrameSourceTypes_Depth,
				&m_pMultiFrameReader);
		}
	}
	if (!m_pKinectSensor || FAILED(hr))
	{
		return E_FAIL;
	}
	UINT16 *depthData = new UINT16[424 * 512];
	IDepthFrameReference* m_pDepthFrameReference = nullptr;
	IColorFrameReference* m_pColorFrameReference = nullptr;
	IDepthFrame* m_pDepthFrame = nullptr;
	IColorFrame* m_pColorFrame = nullptr;
	IMultiSourceFrame* m_pMultiFrame = nullptr;
	ICoordinateMapper* m_pCoordinateMapper = nullptr;
	int count = 0;
	while (count <= 30)
	{
		Sleep(5000);
		while (true)
		{
			hr = m_pMultiFrameReader->AcquireLatestFrame(&m_pMultiFrame);
			if (FAILED(hr) || !m_pMultiFrame)
			{
				continue;
			}
			break;
		}
		if (SUCCEEDED(hr))
			hr = m_pMultiFrame->get_DepthFrameReference(&m_pDepthFrameReference);
		if (SUCCEEDED(hr))
			hr = m_pDepthFrameReference->AcquireFrame(&m_pDepthFrame);
		hr = m_pKinectSensor->get_CoordinateMapper(&m_pCoordinateMapper);
		pcl::PointCloud<MyPointDataType>::Ptr cloud(new pcl::PointCloud<MyPointDataType>);
		cloud->width = 512 * 424;
		cloud->height = 1;
		cloud->is_dense = false;
		cloud->points.resize(cloud->width * cloud->height);
		if (SUCCEEDED(hr))
		{
			hr = m_pDepthFrame->CopyFrameDataToArray(424 * 512, depthData);
			CameraSpacePoint* m_pCameraCoordinates = new CameraSpacePoint[512 * 424];
			hr = m_pCoordinateMapper->MapDepthFrameToCameraSpace(512 * 424, depthData, 512 * 424, m_pCameraCoordinates);
			for (int i = 0; i < 512 * 424; i++)
			{
				CameraSpacePoint cameraP = m_pCameraCoordinates[i];
				if (cameraP.X != -std::numeric_limits<float>::infinity() && cameraP.Y != -std::numeric_limits<float>::infinity() && cameraP.Z != -std::numeric_limits<float>::infinity())
				{
					float cameraX = static_cast<float>(cameraP.X);
					float cameraY = static_cast<float>(cameraP.Y);
					float cameraZ = static_cast<float>(cameraP.Z);
					cloud->points[i].x = cameraX;
					cloud->points[i].y = cameraY;
					cloud->points[i].z = cameraZ;
				}
			}
		}
		string s = "pcd文件名称";
		s += ".pcd";
		pcl::io::savePCDFile(s, *cloud, false);
		std::cerr << "Saved " << cloud->points.size() << " data points." << std::endl;
		s.clear();
		boost::shared_ptr<pcl::visualization::PCLVisualizer> viewer(new pcl::visualization::PCLVisualizer("3D viewer"));
		viewer->addPointCloud(cloud);
		viewer->resetCamera();
		viewer->addCoordinateSystem(0.1);
		viewer->initCameraParameters();
		while (!viewer->wasStopped()){
			viewer->spinOnce();
		}
		count++;
		cout << "test" << endl;
		SafeRelease(m_pDepthFrame);
		SafeRelease(m_pDepthFrameReference);
		SafeRelease(m_pColorFrame);
		SafeRelease(m_pColorFrameReference);
		SafeRelease(m_pMultiFrame);
	}
	m_pKinectSensor->Close();
	SafeRelease(m_pKinectSensor);
	std::system("pause");
	return 0;
}