pierrelouisl/mullan_item_checker

AI Model using YOLO (You Only Look Once) to detect items and check on the data base which of them are needed

Mullan Item Checker

AI Model using YOLO (You Only Look Once) to detect items and check on the data base which of them are needed.

Installation

You will need to use pip to the item checker. It has a few depedencies to use YOLO. You will need to use TrainYourOwnYOLO, here is the steps to install everything (assuming you have Python installed):

git clone https://github.com/AntonMu/TrainYourOwnYOLO
cd TrainYourOwnYOLO/

On (mac/linux) environment

python3 -m venv env
source env/bin/activate

On Windows environment

You will need to open a PowerShell (might need to be admin in order to create the virtual environment.

py -m venv env
.\env\Scripts\activate

Installing packages for Windows/Linux/Mac

pip install pip --upgrade
pip install -r requirements.txt
pip install Pillow pyodbc opencv-contrib-python

You will then need to download the files from this github repository and put them into "TrainYourOwnYOLO\3_Inference". You should have the basic Detector.py file provided by TrainYourOwnYOLO and the two files from this repository.

It is strongly advised to use CuDNN/CUDA for faster implementation, but check the compatibility first. We won't go thru the installation of CUDA here because it would be too long.

Usage

You will need to change line 174 - 177 and 192 - 201 to specify the credentials.

To run the program you will need to go in the folder where the code is situated using the cd command and just run it by using py Semi then press TAB to have the correct path name. You will have to follow TrainYourOwnYOLO README.md to create your own model and use it with this app.

Hikvision program

This is a small program to test the hikvision camera and see it working, or not. You just have to specify the credentials on line 10 of the python code and run it.