Related Experiment Video
Updated: Jun 24, 2026
13:44
Detection of Architectural Distortion in Prior Mammograms via Analysis of Oriented Patterns
Published on: August 30, 2013
Study on Huber fractal image compression.
Jyh-Horng Jeng1, Chun-Chieh Tseng, Jer-Guang Hsieh
1Department of Information Engineering, I-Shou University Kaohsiung County, Taiwan 840, R.O.C. jjeng@isu.edu.tw
Summary
This study introduces Huber fractal image compression (HFIC), a robust method for noisy images. Particle swarm optimization (PSO) significantly reduces HFIC
Area of Science:
- Digital Image Processing
- Computer Vision
- Robust Statistics
Background:
- Fractal Image Compression (FIC) is an efficient lossy compression technique.
- Standard FIC methods can be sensitive to noise and outliers in images.
- Robustness against image corruption is crucial for practical applications.
Purpose of the Study:
- To develop a novel similarity measure for robust fractal image compression.
- To introduce Huber fractal image compression (HFIC) for noise insensitivity.
- To enhance HFIC efficiency using particle swarm optimization (PSO).
Main Methods:
- Embedding linear Huber regression from robust statistics into FIC encoding.
- Developing a robust fractal image compression scheme (HFIC).
- Utilizing particle swarm optimization (PSO) to accelerate the HFIC encoding process.
Main Results:
- The proposed HFIC demonstrates robustness against outliers in corrupted images.
- Particle swarm optimization (PSO) effectively reduces the computational cost of HFIC.
- Image quality is maintained while significantly decreasing encoding time.
Conclusions:
- HFIC offers a robust alternative to standard FIC, particularly for noisy images.
- PSO integration addresses the computational complexity challenge of HFIC.
- The combined approach provides an efficient and resilient image compression solution.
