Related Experiment Video
Updated: Oct 19, 2025
11:00
Experimental Investigation of Secondary Flow Structures Downstream of a Model Type IV Stent Failure in a 180° Curved Artery Test Section
Published on: July 19, 2016
11.7K
A Low Redundancy Wavelet Entropy Edge Detection Algorithm
Yiting Tao1, Thomas Scully2, Asanka G Perera1
1UniSA STEM, Mawson Lakes Campus, University of South Australia, Adelaide, SA 5095, Australia.
Journal of Imaging
|September 26, 2021
Summary
We developed a novel edge detection algorithm using wavelet transform, Shannon entropy, and thresholding. This fast, simple method is noise-resilient and ideal for real-time computer vision applications.
Area of Science:
- Computer Vision
- Image Processing
- Signal Processing
Background:
- Edge detection is a fundamental step in computer vision.
- Efficient edge detection is crucial for real-time image processing applications.
- Existing methods may lack speed, simplicity, or noise resilience.
Purpose of the Study:
- To propose a new, fast, and simple edge detection algorithm.
- To leverage wavelet transform and Shannon entropy for enhanced edge detection.
- To evaluate the algorithm's performance against established methods.
Main Methods:
- Developed a novel algorithm combining wavelet transform, Shannon entropy, and thresholding.
- Utilized Shannon entropy to measure global image structure at each wavelet decomposition level.
- Mathematically formulated the algorithm and conducted comparative analysis.
Main Results:
- The proposed algorithm demonstrates low redundancy.
- The method exhibits significant resilience to noise.
- Achieved performance suitable for real-time image processing.
Conclusions:
- The new algorithm offers an efficient solution for edge detection.
- Its combination of techniques provides robustness and speed.
- Well-suited for various real-time computer vision tasks.
