Skip to content
Permalink
master
Switch branches/tags
Go to file
 
 
Cannot retrieve contributors at this time

How to use OpenDataCam v3 without docker

1. Install dependencies

For Jetsons: Flash jetson to jetpack 4.3+

https://developer.nvidia.com/embedded/jetpack

For GNU/Linux x86_64 machine with a CUDA compatible GPU: Install nvidia drivers and CUDA

https://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html

2. Install Darknet (Neural network framework running YOLO)

Get the source files

NB: Make sure you reinstall darknet entirely if you were on ODC v2.x, for v3 the version has changed.

git clone --depth 1 -b odc https://github.com/opendatacam/darknet

#NB: the changes from https://github.com/alexeyab/darknet are documented here : https://github.com/opendatacam/darknet/pull/6

Modify the Makefile before compiling

Open the Makefile in the darknet folder and make these changes:

For Jetson Nano

# Set these variable to 1:
GPU=1
CUDNN=1
OPENCV=1

# Uncomment the following line
# For Jetson TX1, Tegra X1, DRIVE CX, DRIVE PX - uncomment:
ARCH= -gencode arch=compute_53,code=[sm_53,compute_53]

# Replace NVCC path
NVCC=/usr/local/cuda/bin/nvcc

For Jetson TX2

# Set these variable to 1:
GPU=1
CUDNN=1
OPENCV=1

# Uncomment the following line
# For Jetson Tx2 or Drive-PX2 uncomment
ARCH= -gencode arch=compute_62,code=[sm_62,compute_62]

For Jetson Xavier

# Set these variable to 1:
GPU=1
CUDNN=1
CUDNN_HALF=1
OPENCV=1

# Uncomment the following line
# Jetson XAVIER
ARCH= -gencode arch=compute_72,code=[sm_72,compute_72]

For Generic Ubuntu machine with CUDA GPU

Make sure you have CUDA installed:

# Type this command
nvcc --version

# If it returns Command 'nvcc' not found , you need to install cuda properly: https://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html#package-manager-installation and also add cuda to your PATH with the post install instructions: https://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html#post-installation-actions

Make change to Makefile:

# Set these variable to 1:
GPU=1
CUDNN=1
OPENCV=1

Compile darknet

# Go to darknet folder
cd darknet 
# Optional: put jetson in performance mode to speed up things
sudo nvpmodel -m 0
sudo jetson_clocks
# Compile
make

If you have an error "nvcc not found" on Jetson update path to NVCC in Makefile

NVCC=/usr/local/cuda/bin/nvcc

Download weight file

The .weights files that need to be in the root of the /darknet folder

cd darknet #if you are not already in the darknet folder

# YOLOv4-tiny
wget https://github.com/AlexeyAB/darknet/releases/download/darknet_yolo_v4_pre/yolov4-tiny.weights --no-check-certificate
# YOLOv4
wget https://github.com/AlexeyAB/darknet/releases/download/darknet_yolo_v3_optimal/yolov4.weights --no-check-certificate

Or if you want to copy them over SSH from your main machine with scp

scp yolov4-tiny.weights nvidia@192.168.1.210:/home/nvidia/Documents/darknet

(Optional) Test darknet

# Go to darknet folder
cd darknet 
# Run darknet (yolo) on webcam
./darknet detector demo cfg/coco.data cfg/yolov4-tiny.cfg yolov4-tiny.weights "v4l2src ! video/x-raw, framerate=30/1, width=640, height=360 ! videoconvert ! appsink" -ext_output -dont_show

# Run darknet on file
./darknet detector demo cfg/coco.data cfg/yolov4-tiny.cfg yolov4-tiny.weights opendatacam_videos/demo.mp4 -ext_output -dont_show

# Run darknet on raspberrycam
./darknet detector demo cfg/coco.data cfg/yolov4-tiny.cfg yolov4-tiny.weights "nvarguscamerasrc ! video/x-raw(memory:NVMM),width=1280, height=720, framerate=30/1, format=NV12 ! nvvidconv ! video/x-raw, format=BGRx, width=640, height=360 ! videoconvert ! video/x-raw, format=BGR ! appsink" -ext_output -dont_show

3. Install node.js, mongodb

# Install node.js
sudo apt-get install curl
curl -sL https://deb.nodesource.com/setup_12.x | sudo -E bash -
sudo apt-get install -y nodejs

Mongodb for Jetson devices (ARM64):

# Install mongodb

# Detailed doc: https://computingforgeeks.com/how-to-install-latest-mongodb-on-ubuntu-18-04-ubuntu-16-04/
# NB: at time of writing this guide, we install the mongodb package for ubuntu 16.04 as the arm64 version of it isn't available for 18.04
sudo apt-key adv --keyserver hkp://keyserver.ubuntu.com:80 --recv 9DA31620334BD75D9DCB49F368818C72E52529D4
echo "deb [ arch=amd64,arm64 ] https://repo.mongodb.org/apt/ubuntu xenial/mongodb-org/4.0 multiverse" | sudo tee /etc/apt/sources.list.d/mongodb-org-4.0.list
sudo apt-get update
sudo apt-get install -y openssl libcurl3 mongodb-org

# Start service
sudo systemctl start mongod

# Enable service on boot
sudo systemctl enable mongod

Mongodb for Generic Ubuntu machine with CUDA GPU:

# Install mongodb

# Detailed doc: https://docs.mongodb.com/manual/tutorial/install-mongodb-on-ubuntu/
sudo apt-key adv --keyserver hkp://keyserver.ubuntu.com:80 --recv 9DA31620334BD75D9DCB49F368818C72E52529D4 && \
    echo "deb [ arch=amd64 ] https://repo.mongodb.org/apt/ubuntu xenial/mongodb-org/4.0 multiverse" | tee /etc/apt/sources.list.d/mongodb-org-4.0.list
sudo apt-get update && apt-get install -y --no-install-recommends openssl libcurl3 mongodb-org

# Start service
sudo systemctl start mongod

# Enable service on boot
sudo systemctl enable mongod

4. Install opendatacam

  • Download source
git clone --depth 1 https://github.com/opendatacam/opendatacam.git
cd opendatacam
  • Change "MONGODB_URL" in opendatacam/config.json (default is the docker service mongo URL)
{
  "MONGODB_URL": "mongodb://127.0.0.1:27017"
}
  • Specify ABSOLUTE PATH_TO_YOLO_DARKNET path in opendatacam/config.json
{
  "PATH_TO_YOLO_DARKNET" : "/home/nvidia/darknet"
}
# To get the absolute path, go the darknet folder and type
pwd .
  • Specify VIDEO_INPUT and NEURAL_NETWORK in opendatacam/config.json

For Jetson Nano (recommanded)

{
  "VIDEO_INPUT": "usbcam",
  "NEURAL_NETWORK": "yolov4-tiny"
}

For Jetson Xavier (recommanded)

{
  "VIDEO_INPUT": "usbcam",
  "NEURAL_NETWORK": "yolov4"
}

Learn more in the config documentation page.

  • Install OpenDataCam
cd <path/to/open-data-cam>
npm install
npm run build
  • Run OpenDataCam
cd <path/to/open-data-cam>
npm run start
  • (optional) Config OpenDataCam to run on boot
# install pm2
npm install pm2 -g |

# go to opendatacam folder
cd <path/to/open-data-cam>
# launch pm2 at startup
# this command gives you instructions to configure pm2 to
# start at ubuntu startup, follow them
sudo pm2 startup

# Once pm2 is configured to start at startup
# Configure pm2 to start the Open Traffic Cam app
sudo pm2 start npm --name "opendatacam" -- start
sudo pm2 save
  • (optional) Open ports 8080 8090 and 8070 to outside world on cloud deployment machine
sudo ufw allow 8080
sudo ufw allow 8090
sudo ufw allow 8070

(Optional) How to compile Opencv with Gstreamer support on desktop

sudo apt-get install -y libgstreamer1.0-0 gstreamer1.0-plugins-base gstreamer1.0-plugins-good gstreamer1.0-plugins-bad gstreamer1.0-plugins-ugly gstreamer1.0-libav gstreamer1.0-doc gstreamer1.0-tools libgstreamer1.0-dev libgstreamer-plugins-base1.0-dev

sudo apt-get install -y  pkg-config zlib1g-dev libwebp-dev libtbb2 libtbb-dev libgtk2.0-dev pkg-config libavcodec-dev libavformat-dev libswscale-dev cmake libv4l-dev

sudo apt-get install -y autoconf autotools-dev build-essential gcc git

sudo apt-get install -y ffmpeg

git clone --depth 1 -b 4.1.1 https://github.com/opencv/opencv.git

cd opencv
mkdir build
cd build
# Note here you need to set both FFMPEG and GSTREAMER to ON
# Running this command should output a summary of which dependencies are gonna be build with opencv
# Double check that both gstreamer and ffmpeg are ON
cmake -D CMAKE_INSTALL_PREFIX=/usr/local CMAKE_BUILD_TYPE=Release -D WITH_GSTREAMER=ON -D WITH_GSTREAMER_0_10=OFF -D WITH_CUDA=OFF -D WITH_TBB=ON -D WITH_LIBV4L=ON WITH_FFMPEG=ON -DOPENCV_GENERATE_PKGCONFIG=ON ..

sudo make install

#reload if opencv already installed
sudo /bin/bash -c 'echo "/usr/local/lib" >> /etc/ld.so.conf.d/opencv.conf'
sudo ldconfig