Ecosyste.ms: Awesome
An open API service indexing awesome lists of open source software.
https://github.com/siddeshsambasivam/canny-edge-detector
An implementation of canny edge detection algorithm with sub-pixel accuracy for EE4208.
https://github.com/siddeshsambasivam/canny-edge-detector
Last synced: 22 days ago
JSON representation
An implementation of canny edge detection algorithm with sub-pixel accuracy for EE4208.
- Host: GitHub
- URL: https://github.com/siddeshsambasivam/canny-edge-detector
- Owner: SiddeshSambasivam
- Created: 2022-03-01T06:35:19.000Z (over 2 years ago)
- Default Branch: main
- Last Pushed: 2022-04-09T15:36:13.000Z (over 2 years ago)
- Last Synced: 2024-10-23T03:25:19.388Z (23 days ago)
- Language: Python
- Size: 44.8 MB
- Stars: 3
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
Canny edge detector
================================
The repository is a contains the implementation of canny edge detection algorithm with sub-pixel accuracy.
>This is a project work for the course EE4208: Intelligent system design
## Table of contents
1. [Installation](#installation)
2. [Examples](#examples)
3. [Project requirements](#project-requirements)
4. [Canny edge detection algorithm](#canny-edge-detection-algorithm)
5. [Development setup](#development-setup)
6. [References](#references)
## Installation
### From source
```bash
>> git clone https://github.com/SiddeshSambasivam/Canny-edge-detector.git
>> cd Canny-edge-detector
>> pip install .
```
## Examples
```python
import cv2
from canny_edge_detector import canny, ImageHandler
img = cv2.imread('./tests/data/test.png', 0)
out = canny(
img,
low=5, high=20,
verbose=True,
save_outputs=True
)
```
## Project requirements
1. Take pictures of a B&W chessboard at a high resolution (eg. 3000x4000)
2. Apply your Canny edge detector to these images (grey scale) with accuracy to pixel level (integer)
3. Reduce the size to ¼ of the original image (eg 750x1000), then apply the Canny edge detector with sub-pixel accuracy.
* The output is a list of edge pixels with decimal points numbers.
* Casting these edge pixels to its original size of 3000x4000 and make a comparison with the results obtained in 2.
* Discrepancies can be presented in the report/during the face-to-face assessment.
4. Repeat the above to photos with daily life settings.
## Canny edge detection algorithm
The process of Canny edge detection algorithm can be broken down to five different steps:
1. Apply Gaussian filter to smooth the image to eliminate noise.
* The larger the width of the Gaussian mask, the lower is the detector's sensitivity to noise
* Convolve the filter over the image
2. Find the intensity gradients of the image; image gradient to highlight regions with high spatial derivatives
3. Apply gradient magnitude thresholding or lower bound cut-off suppression to get rid of spurious response to edge detection
4. Apply double threshold to determine potential edges
5. Track edge by hysteresis: Finalize the detection of edges by suppressing all the other edges that are weak and not connected to strong edges.
## Development setup
1. Setup the dev environment
*Prereq: Miniconda was used to setup the development environment.*
```bash
> conda create --name canny python=3.8
> conda activate canny
> pip install -r requirements.txt
```
2. Run the tests
```bash
> pytest -v
```
## References
1. [Canny Edge Detector, Wikipedia](https://en.wikipedia.org/wiki/Canny_edge_detector)
2. [Bill Green, Canny Edge Detection Tutorial, Drexel Autonomous Systems Lab., 2002.](http://masters.donntu.org/2010/fknt/chudovskaja/library/article5.htm#:~:text=Based%20on%20these%20criteria%2C%20the,the%20maximum%20(nonmaximum%20suppression)).