vision_opencv

NOTE: LOOK vision_opencv_bridge.txt for more info about installation

INSTALLATION

cd ~/quadcopter_drone_ws/src
git clone https://github.com/ros-perception/vision_opencv.git
cd vision_opencv
git checkout humble
cd ~/quadcopter_drone_ws
colcon build --packages-select cv_bridge --cmake-args -DOpenCV_DIR=/usr/local/lib/cmake/opencv4

Note: Replace /path/to/opencv4.11/cmake with the actual path to OpenCV 4.11's CMake configuration files (e.g., /usr/local/lib/cmake/opencv4)

source install/setup.bash

Verify Installation

After building, ensure the OpenCV version linked by cv_bridge is correct. You can check by inspecting the logs or running the cv_bridge node to see if it recognizes the installed OpenCV version.

If everything builds and links correctly, you can proceed to build your yolo_opencv package.

Verify Integration with Your yolo_opencv Package

Ensure cv_bridge builds successfully. After building, you can build your ros2 package:

colcon build --packages-select <ros2_package_name> --cmake-args -DOpenCV_DIR=/usr/local/lib/cmake/opencv4

This setup ensures that the cv_bridge package from vision_opencv is properly integrated with your workspace and uses your custom OpenCV installation.

=============

ros2 vision_opencv contains packages to interface ROS 2 with OpenCV which is a library designed for computational efficiency and strong focus for real time computer vision applications. This repository contains:

• cv_bridge: Bridge between ROS 2 image messages and OpenCV image representation
• image_geometry: Collection of methods for dealing with image and pixel geometry
• opencv_tests: Integration tests to use the capability of the packages with opencv
• vision_opencv: Meta-package to install both cv_bridge and image_geometry

In order to use ROS 2 with OpenCV, please see the details within cv_bridge package.