opencv支持红外结构光相机吗 opencv gige相机

最近在做无人驾驶方面的东西，老师给了一个摄像头，是Gige协议，公司叫microview的不知名公司，连驱动也只有ubuntu16.04的。百度一搜，有关ROS的都是USB摄像头，仔细想想，USB摄像头还是小儿科了一点，性能应该不如千兆网口的。基本现在市面买的摄像头也没提供ros方案。于是只能自己写package。

用千兆网口摄像机的几种解决方案

1千兆网口，但协议是rstp之类的网络摄像头。这种摄像头是输出h.265或者MJPEG之类的格式。不管在opencv还是ros都容易调用
2 gige/gig version协议摄像头。这是本文重点。
如果买的是巴斯勒(basler)或者prosilica牌子的摄像头,你可以：

1 使用现有包
basler有官方的ros包，名称叫做pylon-ros-camera。这是专门为basler品牌设计的ros package。同时，basler官网可以下载pylon软件，功能界面都不错，方便调参。prosilica_driver是prosilica相机的package。经过测试，我的相机不能被它们识别。

当你买的是其他品牌时候：
首先我找到了camera_aravis这个ros package。相当不错，基本支持了所有的摄像机协议，同时默认发布两个topic: /camera/image_raw和/camera/camera_info。
其实这时候就完全可以在ros下使用了。可是在rviz的image界面订阅/camera/image_raw的时候，图像黑屏，奇怪的是一直在显示接收到图片。最终发现，camera_aravis自动将增益设置为1。正常情况下，增益不应该低于15。

进一步改进

此时，你应该已经能在rviz下看到接收到的图像了。但你发现是黑白的，为了有更好的效果，以及为了后期使用视觉算法，我决定将其转换成OPENCV的 MAT图像格式。
一开始我想用image_proc包实现，但始终报错，

编译自己的ROS包

我考虑过对每个像素进行操作，通过指针的方式，处理bayergr8原始图像数据，但始终有core dump的报错（问题可能出在没有动态分配内存）。总之，我决定使用opencv自带的转换参数：cv::COLOR_Bayer2RGB即可。
代码如下：
main.cpp

#include <ros/ros.h>
#include <image_transport/image_transport.h>
#include <cv_bridge/cv_bridge.h>
#include <sensor_msgs/image_encodings.h>
#include <opencv2/imgproc/imgproc.hpp>
#include <opencv2/highgui/highgui.hpp>
#include "std_msgs/String.h"
#include<vector>
#include<iostream>
#include <opencv2/opencv.hpp>

//
/// \brief cxg_bgr
/// \param array_bayer
///this is convert of raw bayerrg8 to rgb format(1280*1024)
/
void cxg_bgr(unsigned char *array_bayer)
{
    cv::Mat opencv_Bayer(cv::Size(1280,1024),CV_8UC1);  // bayer data in opencv (value is equl to raw ros bayer data)
    cv::Mat opencv_Bgr(cv::Size(1280,1024),CV_8UC3);    //final bgr data in opencv
    int framesize = 1280 * 1024;
    unsigned char * bayerdata=( unsigned char *) malloc (sizeof( unsigned char * ) * framesize);
    bayerdata=array_bayer;    //bayerdata is a temp variable
    opencv_Bayer.data=bayerdata;
    printf("opencv bayer: %d\n",opencv_Bayer.data[10000]);
    cv::cvtColor(opencv_Bayer, opencv_Bgr, cv::COLOR_BayerGR2RGB);
    printf("opencv bgr channel1:%d channel2:%d  channel3:%d\n",opencv_Bgr.at<cv::Vec3b>(1000,1000)[0],opencv_Bgr.at<cv::Vec3b>(1000,1000)[1],opencv_Bgr.at<cv::Vec3b>(0,0)[2]);
    imshow("view",opencv_Bgr);
    cv::waitKey(30);
}

void convertCallback(const sensor_msgs::ImageConstPtr& msg)
{

   // unsigned char *data;
    try
    {

         //ROS_INFO("%d",msg->step);
         //  std::vector<unsigned char> dataccc;
         const unsigned char * tmp = &(msg->data[0]);
         unsigned char array[1310720];
         unsigned char *array_bayer = (unsigned char *)tmp;
         unsigned char *array_bgr;
         cxg_bgr(array_bayer);

    }
    catch (cv_bridge::Exception& e)
    {
      ROS_ERROR("Could not convert from '%s' to 'bgr8'.", msg->encoding.c_str());
    }

}


int main(int argc, char** argv)
{

  ros::init(argc, argv, "cxg_camera");

  ros::NodeHandle cxg;
  cv::namedWindow("view");
  cv::startWindowThread();
  image_transport::ImageTransport it(cxg);
  image_transport::Subscriber sub = it.subscribe("/camera/image_raw", 1000, convertCallback);
  ros::spin();
  return 0;
}

CMakelist.txt

# http://ros.org/doc/groovy/api/catkin/html/user_guide/supposed.html
cmake_minimum_required(VERSION 2.8.3)
project(cxg_camera)
set(CMAKE_MODULE_PATH ${CMAKE_MODULE_PATH} ${CMAKE_CURRENT_SOURCE_DIR}/CMakeModules)
find_package(OpenCV REQUIRED)
find_package(catkin REQUIRED COMPONENTS roscpp sensor_msgs image_transport camera_info_manager dynamic_reconfigure driver_base tf dynamic_reconfigure cv_bridge std_msgs)
catkin_package(
    CATKIN_DEPENDS roscpp sensor_msgs image_transport camera_info_manager dynamic_reconfigure driver_base tf  cv_bridge std_msgs
    INCLUDE_DIRS
    LIBRARIES
)

include_directories(include ${catkin_INCLUDE_DIRS} ${OpenCV_INCLUDE_DIRS} )
add_executable(cxg main.cpp)
target_link_libraries(cxg ${catkin_LIBRARIES} ${OpenCV_LIBS})
add_dependencies(cxg ${PROJECT_NAME}_gencfg).

packae.xml

<package>
  <name>cxg_camera</name>
  <version>2.0.0</version>
  <description>camera_aravis: An ethernet camera driver for ROS.</description>

  <maintainer email="[e-mail]">Steve Safarik</maintainer>

  <author email="[e-mail]">Steve Safarik</author>
  <author email="[e-mail]">Andrew Straw</author>

  <license>LGPLv2</license>

  <url type="website">http://github.com/ssafarik/camera_aravis</url>
  
  <author>strawlab</author>
  <author>Steve Safarik</author>

  <buildtool_depend>catkin</buildtool_depend>
  
  <build_depend>roscpp</build_depend>
  <build_depend>sensor_msgs</build_depend>
  <build_depend>image_transport</build_depend>
  <build_depend>camera_info_manager</build_depend>
  <build_depend>dynamic_reconfigure</build_depend>
  <build_depend>driver_base</build_depend>
  <build_depend>tf</build_depend>
  <build_depend>cv_bridge</build_depend>
  <build_depend>std_msgs</build_depend>
<build_depend>opencv2</build_depend>
  

  
  <run_depend>roscpp</run_depend>
  <run_depend>sensor_msgs</run_depend>
  <run_depend>image_transport</run_depend>
  <run_depend>camera_info_manager</run_depend>
  <run_depend>dynamic_reconfigure</run_depend>
  <run_depend>driver_base</run_depend>
  <run_depend>tf</run_depend>
  <run_depend>cv_bridge</run_depend>
  <run_depend>std_msgs</run_depend>
<run_depend>opencv2</run_depend>

</package>

最终效果：