Highly Effective GPU Realization of Discrete Wavelet Transform for Big-Data Problems

Puchala, Dariusz; Stokfiszewski, Kamil

doi:10.1007/978-3-030-77961-0_19

Cited by 3 publications

(3 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The one-dimensional discrete wavelet transform (DWT1D) in the practical tasks of digital signal processing and analysis is implemented in most cases as a two-channel bank of filters with the structure depicted in Fig. 1 (see [15][16][17][18][19]). When using the polyphase notation an input signal can be described in the following form X…”

Section: One-dimensional Orthogonal Wavelet Transform and Lattice Str...mentioning

confidence: 99%

“…In such cases it is required to describe wavelet transform with a parametric model that can be characterized by nonredundant number of parameters. In this case the effective tool for calculation of one-dimensional (and also two-dimensional by the means of row-column approach) orthogonal transforms are lattice structures which, not only allow for the reduction of the number of additions and multiplications, but also can be characterized by accurate and non-redundant parametrization (see [15][16][17][18][19][20][21]).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Bulletin of the Polish Academy of Sciences: Technical Sciences

Puchala

2022

View full text Add to dashboard Cite

Discrete two-dimensional orthogonal wavelet transforms find applications in many areas of analysis and processing of digital images. In a typical scenario the separability of two-dimensional wavelet transforms is assumed and all calculations follow the row-column approach using one-dimensional transforms. For the calculation of one-dimensional transforms the lattice structures, which can be characterized by high computational efficiency and non-redundant parametrization, are often used. In this paper we show that the row-column approach can be excessive in the number of multiplications and rotations. Moreover, we propose the novel approach based on natively two-dimensional base operators which allows for significant reduction in the number of elementary operations, i.e., more than twofold reduction in the number of multiplications and fourfold reduction of rotations. The additional computational costs that arise instead include an increase in the number of additions, and introduction of bit-shift operations. It should be noted, that such operations are significantly less demanding in hardware realizations than multiplications and rotations. The performed experimental analysis proves the practical effectiveness of the proposed approach.

show abstract

Section: One-dimensional Orthogonal Wavelet Transform and Lattice Str...mentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Bulletin of the Polish Academy of Sciences: Technical Sciences

Puchala

2022

View full text Add to dashboard Cite

show abstract

“…Qianjiao Wu et al [29] CUDA algorithm improved the computational efficiency of multiscale DEM analysis, reducing response times.Engels et al [30]'s CUDA-SHAPE algorithm and Axel Davy et al [31] GPU-accelerated denoising solution both enhanced the performance of their respective software and algorithms. The works of Bhaskar Jyoti Borah et al [32] and Dariusz Puchala and Kamil Stokfiszewski [33] further demonstrated the effectiveness of GPU acceleration in image processing.Tianru Xue et al [34] real-time anomaly detection technology and Raghav G. Jha et al [35] TRG accelerated computation method, along with Qiyang Xiong et al [36] PIC simulation optimization scheme, all showcase the potential of GPUs in enhancing remote sensing data processing capabilities. These studies provide new directions for efficient processing of remote sensing data.…”

Section: Introductionmentioning

confidence: 96%

Parallel Acceleration Algorithm for Wavelet Denoising of UAVAGS Data Based on CUDA

Xiong,

Wang,

Qiao

et al. 2024

Preprint

View full text Add to dashboard Cite

The computational efficiency is low when the vast volume of unmanned aerial vehicle airborne gamma-ray spectrum (UAVAGS) data is handled by wavelet denoising in CPU. So, a CUDA-based GPU parallel solution is recommended to resolve this issue in this paper. This proposed solution aims to significantly enhance the efficiency of parallel acceleration for wavelet denoising of UAVAGS data. In the preliminary stage, experiments were conducted with varying block sizes to investigate the influence of different block sizes on processing time. The objective was to identify the most suitable block size for efficiently processing UAVAGS data. Subsequently, a performance evaluation was conducted by comparing the acceleration ratios of GPU and CPU for different data volumes, as well as varying wavelet basis functions under the same data volume conditions. Finally, by intentionally introducing noise, calculations were performed to determine the optimal wavelet basis function concerning signal-to-noise ratio after denoising. The research findings indicate that the optimal two-dimensional block size falls within the range of 64×64 to 128×128. The majority of wavelet basis functions achieved acceleration ratios exceeding 100-fold in total processing time, with the coif5 wavelet basis function reaching an acceleration ratio of 185-fold. Comparative analysis of various denoising functions revealed that, under low signal-to-noise ratios, these functions exhibited insufficient denoising effects, while at high signal-to-noise ratios, there was a risk of excessive denoising. However, significant denoising effects were observed when employing hard thresholding with coif5, soft thresholding, and an improved thresholding method with db3.

show abstract