Solarian PV Module EL Image Deduplication Tools has been written using Python to address the duplicate
image issues at the PV module manufacturers where modules with different serial numbers have the same
EL image. This issue has been arised during our regular factory controls and we've developed a tool
to address it.
Report Bug
·
Request Feature
At Solarian, we are a team of engineers dedicated to offers services of engineering at an exceptional quality. We push our hardest in terms of engineering and analysis.
In order to increase the PV module production quality; we do inspection of photovoltaic modules at the factories. One thing we also check at the factories is duplicate EL images. Some production facilities tend to use bad software that creates/uses same EL image for different serial numbers. In order to spot those issues we've developed the following code. This software simply checks for similar images using pre-trained MobileNet CNN.
Please feel free to fork or send pull requests. Please keep the code as minimal as possible.
This project has been coded with Python 3 using Tensorflow, imagededup and nmslib.
Follow the steps below to prepare the environment for the project.
First you need to get Python 3 installed and running with dependencies correctly installed.
- bash
sudo apt update
sudo apt-get -y dist-upgrade
sudo apt-get -y install git python3-distutils python3-dev build-essential
sudo curl https://bootstrap.pypa.io/get-pip.py -o get-pip.py
sudo python3 get-pip.py
sudo pip3 install tensorflow imagededup pillow keras
- Clone the repo (Change the directory if you want)
git clone https://github.com/orcunbaslak/pvmodule-el-deduplicator /home/pi/pvmodule-el-deduplicator
- Edit the python file
deduplicate.py
and provide the path to the folder which holds the EL images.
nano deduplicate.py
You can feed the file to Python3 interpreter and it's all good to go given you've entered the correct folder path. Please do check the comments in the python file.
python3 deduplicate.py
See the open issues for a list of proposed features (and known issues).
Contributions are what make the open source community such an amazing place to be learn, inspire, and create. Any contributions you make are greatly appreciated.
- Fork the Project
- Create your Feature Branch (
git checkout -b feature/NewContrib
) - Commit your Changes (
git commit -m 'Add a new contribution'
) - Push to the Branch (
git push origin feature/NewContrib
) - Open a Pull Request
Distributed under the MIT License. See LICENSE
for more information.
Orçun Başlak - @orcunbaslak - website - [email protected]
Solarian Enerji - @solarianenerji - website - [email protected]
Project Link: https://github.com/orcunbaslak/pvmodule-el-deduplicator