LidarMeasure/locate/cnn3d_segmentation.h

#ifndef __3DCNN__LOCATER__HH_
#define __3DCNN__LOCATER__HH_
#include "opencv/highgui.h"
#include "opencv2/opencv.hpp"
#include <pcl/point_types.h>
#include <pcl/common/common.h>
#include <string>
#include "../error_code/error_code.h"

/*
 * 3dcnn网络识别车两中轮胎点
 */
class  Cnn3d_segmentation
{
public:
	Cnn3d_segmentation(int l,int w,int h,int freq,int nClass);
	//设置3dcnn网络参数
    Error_manager set_parameter(int l, int w, int h,int freq,int classes);
	virtual ~Cnn3d_segmentation();
	//初始化网络参数
	//weights:参数文件
	virtual Error_manager init(std::string weights);

	//预测
	//cloud:输入点云
	//rect:输出旋转矩形框
	//save_dir:中间文件保存路径
	virtual Error_manager predict(pcl::PointCloud<pcl::PointXYZ>::Ptr cloud,
	    float& center_x,float& center_y,
	    float& wheel_base,float& width,float& angle, std::string save_dir);

protected:
    //根据设置的参数将点云转换成网络输入数格式
	float* generate_tensor(pcl::PointCloud<pcl::PointXYZ>::Ptr cloud, float min_x, float max_x,
		float min_y, float max_y, float min_z, float max_z);
    //将识别结果转换成点云
	std::vector<pcl::PointCloud<pcl::PointXYZRGB>::Ptr> decodeCloud(pcl::PointCloud<pcl::PointXYZ>& cloud,
		float* data, float min_x, float max_x, float min_y, float max_y, float min_z, float max_z);
protected:
	//判断矩形框是否找到
	//points:输入轮子中心点, 只能是3或者4个
	Error_manager isRect(std::vector<cv::Point2f>& points);
	//kmeans算法聚类,拆分四个轮子,返回四轮中心
	//cloud:输入点云
    std::vector<pcl::PointCloud<pcl::PointXYZ>::Ptr> kmeans(
        pcl::PointCloud<pcl::PointXYZRGB>::Ptr cloud, std::string cluster_file_path);
	bool check_box(cv::RotatedRect& box, pcl::PointCloud<pcl::PointXYZ>::Ptr cloud);
	///IIU判断
	bool check_IOU(cv::RotatedRect& box, pcl::PointCloud<pcl::PointXYZ>::Ptr cloud);
    /*
     * 对点云作pca分析,然后在最小轴方向过滤, 防止轮胎左右两边出现噪声
     * cloud_in:输入点云
     * cloud_out:输出点云
     */
    Error_manager pca_minist_axis_filter(pcl::PointCloud<pcl::PointXYZ>::Ptr cloud_in,
                                                             pcl::PointCloud<pcl::PointXYZ>::Ptr cloud_out);

    /*
     *计算两点距离
     */
    static double distance(cv::Point2f p1, cv::Point2f p2);
protected:
	int m_lenth;
	int m_width;
	int m_height;
	int m_freq;
	int m_nClass;
};
#endif