Lidar_Projection

OS: Ubuntu 18.04 LTS

Goals

Project lidar points to rgb images

Environment

• ROS Melodic

• Miniconda (optional)
  In this project, environment `Thermo' is created

• xterm

sudo apt-get install xterm

• Libraries
  This project requires following packages: Ouster Lidar, HIKRobot Camera.
  Make sure all packages are correctly configured. Detailed hardwardsetting are list below.

Prerequisites

• CUDA CuDNN
• HIKRobot RGB Camera
• Run CV_Bridge in Python3(for ROS Melodic)

Installation

1. clone this project to home folder and init submodule

cd ~/Lidar_Projection
git submodule init
git submodule update
cd src/ouster-ros
git submodule init
git submodule update

2. Install YoloV5

conda activate thermo
cd ~/Thermal_Cognitive/src/yolov5/
git checkout v6.2
pip install -r requirements.txt

3. 2. Install Ouster Lidar dependencies

sudo apt install -y                     \
    ros-$ROS_DISTRO-pcl-ros             \
    ros-$ROS_DISTRO-rviz                \
    ros-$ROS_DISTRO-tf2-geometry-msgs

where $ROS-DISTRO is either melodic or noetic.

Additional dependenices:

sudo apt install -y \
    build-essential \
    libeigen3-dev   \
    libjsoncpp-dev  \
    libspdlog-dev   \
    cmake

4. Build project

cd ~/Lidar_Projection
catkin_make --cmake-args -DCMAKE_BUILD_TYPE=Release

5. Set IP Set your computer IP to 192.168.xx.xx

6. Launch

conda activate thermo
cd ~/Lidar_Projection/shell
./fusion.sh

Settings

• HIKRobot RGB Camera
• YOLOV5
• Ouster Lidar
• CUDA CUDnn

In this project, camera intrinsic calibration jupyter notebook can be found ./calibration/CameraIntrinsic/Camera intrinsic Matrix.ipynp

Camera Lidar calibration please refer to link