Optical-Inspection-of-PCBs

This project is intended to execute optical inspection for bare, assembled boards and solder joints based on template and defective board image comparison. By editing relevant parts and changing segmentation parameters according to which yields the best in your case you can inspect your defective board based on template board image in JPEG format or there are some board image pairs available to test in BOARD LIBRARY file to instantiate the procedure. Please refer to the research paper for more information. https://ieeexplore.ieee.org/document/8936659

Installation

Anaconda

Windows

• Open up an anaconda prompt
• Create a new environment called pcb with python3.6
  • conda create -n pcb python=3.6
• Switch to the new environment
  • conda activate pcb
• Install qt4
  • Download the relevant pyqt4 wheel from here
  • Download PyQt4‑4.11.4‑cp36‑cp36m‑win_amd64.whl if 64 bit or otherwise
  • Go to download folder
  • Build and install the wheel
    • for example: pip install PyQt4‑4.11.4‑cp36‑cp36m‑win_amd64.whl if it is 64 bit one
• Install other packages
  • pip install opencv-python==3.2.0.8 matplotlib numpy pytesseract imutils
• Install Tesseract
  • Download this (Or go to here and download the required version)
  • Install it with default settings

Run

• python InspectonGUI.py