Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/galeone/rods
Visual inspection of motorcycle connecting rods. Improves the result of the watershed algorithm exploiting the domain knowledge
https://github.com/galeone/rods
Last synced: 5 days ago
JSON representation
Visual inspection of motorcycle connecting rods. Improves the result of the watershed algorithm exploiting the domain knowledge
- Host: GitHub
- URL: https://github.com/galeone/rods
- Owner: galeone
- License: other
- Created: 2016-02-09T10:13:50.000Z (almost 9 years ago)
- Default Branch: master
- Last Pushed: 2016-05-28T12:44:06.000Z (over 8 years ago)
- Last Synced: 2024-10-30T06:27:31.871Z (about 2 months ago)
- Language: C++
- Homepage:
- Size: 539 KB
- Stars: 0
- Watchers: 3
- Forks: 1
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Visual inspection of motorcycle connecting rods
Rods analyzes the images in the `rods` folder in order to extract:
1. type
2. baricenter
3. orientation
4. width
5. height
6. length
7. width at the baricenter
8. number of holesFor every hole:
1. the center position
2. the diameterThis is a classic computer vision problem. The peculiarity is how the problem of the touching rods has been solved.
Usually, a border following approach is used. In that case, I used the watershed transform (with markers).
Exploiting the knowledge about the domain, we can improve the markers and make them an approximation of the rod.
In such way the watershed algorithm performs better and the touching rods can easily speared and analyzed singularly.
The idea implementation starts at this line: https://github.com/galeone/Rods/blob/master/main.cpp#L217