Run the tool

OpenCV-Testing is a tool of fuzzing Opencv-python, you can run main.py directly.

standarized API info

you can find in

OpenCV-Testing/API_info.py

Docker and OpenCV Coverage Setup

Follow these steps to build and run a Docker container for OpenCV code coverage testing, and generate a coverage report.

1. Build Dockerfile

   docker build -t opencv-coverage .

2. Run Docker Container

   docker run -it --rm -v "Your_Location\cover_opencv\new_cov\opencv_docker_code_coverages\OpenCV-Testing:/app" --name opencv_coverage_container_1 opencv-coverage

3. Check the Files

   ls -l

4. Run Python Script

   python3 main.py

5. Navigate to OpenCV Build Directory

   cd /usr/local/src/opencv/build/

6. Install gcovr

   pip install gcovr

7. Generate Coverage Report

   gcovr -r /usr/local/src/opencv --html --html-details -o coverage_report.html

8. Find Container ID

   docker ps

9. Copy Coverage Report to Host

   docker cp <container_id>:/usr/local/src/opencv/build/coverage_report.html <path_on_host>

   Example:

   docker cp c4f9932f08ee:/usr/local/src/opencv/build/coverage_report.html .

10. Open Coverage Report

    • macOS:

      open <path_on_host>/coverage_report.html

    • Windows:

      start <path_on_host>\coverage_report.html

    • Linux:

      xdg-open <path_on_host>/coverage_report.html