A MASSIVE PARALLEL COMPUTING ALGORITHM BASED ON GPU/CUDA FOR PROCESSING LARGE-SCALE MULTIFRACTAL IMAGES

Authors

  • J.Sh. Khursanboyev Author

Keywords:

multifractal analysis, box-counting, parallel computing, GPU, NVIDIA CUDA, speedup factor, real-time processing.

Abstract

This paper presents a massive parallel computing algorithm implemented on GPU (NVIDIA CUDA) architecture for the real-time computation of multifractal parameters in high-resolution digital images. The conventional sequential computational complexity of O(NlogN) has been optimized through parallel box-partitioning and dynamic shared memory allocation within CUDA threads. Experimental results demonstrate that the proposed parallel algorithm reduces processing latency by 6.0 to 9.5 times compared to CPU-based execution.

References

1. Potapov A.A. Fractals in Modern Radiophysics and Image Analysis. – Moscow: Radiotekhnika, 2005. – 368 p.

2. Sanders J., Kandrot E. CUDA by Example: An Introduction to General-Purpose GPU Programming. – Addison-Wesley, 2010.

3. Arneodo A., Bacry E., Muzy J.F. The thermodynamics of fractals revisited with wavelets // Physica A. – 1995. – Vol. 213, No. 1-2. – P. 232-275

Published

2026-10-13