Skip to content

README_EN.md

Latest commit

 

History

History
180 lines (109 loc) · 3.98 KB

README_EN.md

File metadata and controls

180 lines (109 loc) · 3.98 KB

[EN/JP]

Camera System

Python Environment

[x64(Windows/Linux)]

  • Python 3.9

  • OpenVINO 2022.2

  • DearPyGUI

  • websocket-client

  • Pillow

  • numpy

  • opencv-python

[Raspberry Pi 3B(Arm) + Intel NCS]

  • Python 3.7

  • OpenVINO 2020.3

  • websocket-client

  • Pillow

  • numpy

  • opencv-python

How to build & run

[Windows (x64, exe version)]

  1. Download the MICS Camera System file (exe version) and extract it to the desired location.

  2. Run the file "mics_camera_system.exe".

[Windows (x64, Python Edition)

  1. Install Python 3.9.

  2. Create a virtual environment of your choice.

  3. Download the MICS Camera System file (Python version) and extract it to an arbitrary location.

  4. Open "resources/platform.jsonc" and rewrite the necessary items according to the instructions in the file.

  5. Activate the virtual environment.

  6. Go to the extracted folder in Terminal.

  7. Install the module by entering the following commands in order.

    python -m pip install --upgrade pip
pip install -r requirements.txt

  8. Run "mics_camera_system.py".

    python mics_camera_system.py

[Arm(Raspberry Pi 3B) + Intel NCS]

  1. Install Python 3.7 . Open a terminal and type the following command to install Python 3.7 .

     sudo apt-get update
 sudo apt-get upgrade -y

 wget https://www.python.org/ftp/python/3.7.13/Python-3.7.13.tgz

 tar zxvf Python-3.7.13.tgz

 cd Python-3.7.13
 ./configure --enable-shared
 make
 sudo make install

 python3.7 -v

  2. Download the OpenVINO Runtime.

    Download l_openvino_toolkit_runtime_raspbian_p_2020.3.194.tgz from storage.openvinotoolkit.org .

  3. Navigate to the folder where you downloaded the software.

  4. Create a destination folder for installation.

    sudo mkdir -p /opt/intel/openvino

  5. Extract the downloaded file.

    sudo tar -xf l_openvino_toolkit_runtime_raspbian_p_2020.3.194.tgz --strip 1 -C /opt/intel/openvino

  6. Updates environment variables. (Must be run each time the terminal is opened)

    source /opt/intel/openvino/bin/setupvars.sh

    (Optional) Automate environment variable updates by writing them to .bash_profile. You do not need to execute the above commands each time if you execute the following commands.

    echo "source /opt/intel/openvino/bin/setupvars.sh" >> ~/.bashrc

  7. Activate Intel NCS.

    sudo usermod -a -G users "$(whoami)"
source /opt/intel/openvino/bin/setupvars.sh
sh /opt/intel/openvino/install_dependencies/install_NCS_udev_rules.sh

  8. Go to any location to create a virtual environment.

  9. Create a virtual environment.

    python3.7 -m venv [Any name]

  10. Activate the virtual environment.

    . [Any name]/bin/activate

  11. Automate the startup of the virtual environment. Open the file "/opt/intel/openvino/bin/setupvars.sh" in your editor.

    sudo nano /opt/intel/openvino/bin/setupvars.sh

    Add the following text to the file and save it.

    command=". [path of a virtual environment]/bin/activate"
eval $command

  12. Download the MICS Camera System file (Python version, for RaspberryPI3+NCS) and extract it to the desired location.

  13. Navigate to the extracted location using a terminal.

  14. Install the module by entering the following commands in order

    python -m pip install --upgrade pip
pip install -r requirements_pi3.txt

  15. Connect the Intel NCS to the Raspberry Pi if it is not already connected.

  16. Run "mics_camera_system.py".

    python mics_camera_system.py

Configuration required for the environment

[Camera Settings]

The resolution, frame rate, and angle of view of the camera must be set. Set these in the GUI or edit "resources/settings.json".