To install OpenCV with CUDA support on your NVIDIA Jetson Nano, follow these steps:
Step 1: Update and Upgrade the System First, ensure that your system is updated:
sudo apt-get update sudo apt-get upgrade Step 2: Install Dependencies Install the required dependencies:
sudo apt-get install -y build-essential cmake git libgtk2.0-dev pkg-config libavcodec-dev libavformat-dev libswscale-dev python-dev python-numpy libtbb2 libtbb-dev libjpeg-dev libpng-dev libtiff-dev libdc1394-22-dev
Step 3: Download OpenCV and OpenCV Contrib Download the OpenCV and OpenCV Contrib source code:
cd ~ git clone https://github.com/opencv/opencv.git git clone https://github.com/opencv/opencv_contrib.git Step 4: Checkout to a Specific Version (Optional)
You can checkout to a specific version if needed:
cd ~/opencv git checkout 4.1.1 cd ~/opencv_contrib git checkout 4.1.1 Step 5: Create a Build Directory Create a directory for the build process:
cd ~/opencv mkdir build cd build Step 6: Configure the Build with CMake Configure the build to enable CUDA:
cmake -D CMAKE_BUILD_TYPE=Release
-D CMAKE_INSTALL_PREFIX=/usr/local
-D OPENCV_EXTRA_MODULES_PATH=~/opencv_contrib/modules
-D WITH_CUDA=ON
-D ENABLE_FAST_MATH=1
-D CUDA_FAST_MATH=1
-D WITH_CUDNN=ON
-D OPENCV_DNN_CUDA=ON
-D WITH_CUBLAS=1
-D BUILD_opencv_cudacodec=OFF
..
Step 7: Build and Install OpenCV
Start the build process. This can take several hours:
make -j4 # Use the number of cores you have, for Jetson Nano 2GB, use -j2 sudo make install sudo ldconfig Step 8: Verify the Installation To verify that OpenCV is installed with CUDA support, run the following Python script:
import cv2 print(cv2.getBuildInformation()) Look for the lines indicating CUDA support:
yaml Use CUDA: YES Use CUDNN: YES Troubleshooting Ensure you have the correct CUDA version installed: The installed version of CUDA should match the one compatible with your Jetson Nano. The above steps are tailored for CUDA 10.x. Check for any missing dependencies: Ensure all necessary libraries are installed. Review CMake output: During the CMake configuration step, ensure that CUDA is found and enabled in the configuration summary. If you encounter any errors, please share the specific error messages for further assistance.