Extended Multi WLS Method for Lossless Image Coding

Area of Science:

• Computer Science
• Image Processing
• Data Compression

Background:

• Current image lossless coding methods face challenges in balancing data compaction efficiency and computational complexity.
• Existing algorithms often require significant processing time, limiting their practical application.

Purpose of the Study:

• To present a novel, highly efficient image lossless coding method.
• To improve upon existing techniques in terms of both data compaction and processing speed.

Main Methods:

• The proposed method utilizes a cascaded approach based on the Weighted Least Square (WLS) technique.
• Key improvements include a two-step Non-Local Means (NLMS) predictor with Context-Dependent Constant Component Removing.
• Prediction errors are encoded using an efficient binary context arithmetic coder.

Main Results:

• The new algorithm demonstrates superior data compaction compared to current state-of-the-art methods.
• Despite computational complexity, the method achieves better time efficiency than its main competitors.
• Performance was validated on a standard set of benchmark images.

Conclusions:

• The presented cascaded image lossless coding method offers a significant advancement in data compaction and processing efficiency.
• The integration of WLS, advanced NLMS prediction, and arithmetic coding provides a robust solution for image compression.