limsukjing/oring-inspection

A program to inspect images of o-rings for visible defects.

O-ring Inspection

An o-ring is a rubber gasket used in a wide range of ducting and pipework applications, particularly for sealing the junction between two surfaces in liquid or gaseous systems.

This is a simple Python program designed to inspect images of o-rings for visible defects (e.g. chipped or broken rings) that may have been caused during the manufacturing process.

Getting Started

These instructions will get you a copy of the project up and running on your local machine for development purposes.

Prerequisites

If you already have Python 2 or Python 3 installed, you're good to go! If not, install and set up Python 3 on your machine using the packages available here. Have a look at the following links:

• Linux/UNIX
• macOS
• Windows

Installing

Use Git from the command line to clone this repo to your local machine:

$ git clone https://github.com/limsukjing/oring-inspection.git

To run the main script:

python inspection.py 

Algorithm

1. Read the o-ring images from the res folder in a loop.
2. Image segmentation: partition each image into foreground and background using either a histogram-based or a clustering-based thresholding method.
3. Binary morphology: perform dilation and erosion to fill interior holes, or generally remove any imperfections in the images.
4. Connected-component labelling (CCL): extract the foreground pixels, i.e. pixels belonging to the o-ring and assign unique labels to groups of pixels that are connected. Use the labels assigned to remove broken pieces.
5. Contour detection: extract contours and generate a bounding box around the labelled o-ring.
6. Image analysis: analyze the extracted region to determine whether the o-rings are defective and label them accordingly.
   • split the o-ring in half vertically and calculate its thickness, i.e. the number of foreground pixels, on each side.
   • in theory, an o-ring should be symmetrical so its thickness on each side must be approximately equal.
   • the o-ring will be classified as failed as long as it's chipped or broken.
7. Add image processing time as an annotation to the output image.

Built With

• NumPy - The fundamental package for scientific computing with Python, which is used to transform images into arrays that can be manipulated.
• Matplotlib - A comprehensive library for creating static, animated and interactive visualizations in Python.
• OpenCV - An open-source Computer Vision library used only for reading cv2.imread(), displaying cv2.imshow() and annotating cv2.putText() the images. All image processing tasks are done without using OpenCV functions.

Author

Suk Jing Lim - Please email me if you have any questions.

Acknowledgement

4th year Computer Vision project supervised by Dr. Simon McLoughlin.