Cellular Neural Networks simulation on a parallel graphics processing unit

Fernandez, A.; Martín, Rodrigo; Farguell, Enric; Pazienza, Giovanni Egidio

doi:10.1109/cnna.2008.4588679

Cited by 12 publications

(8 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Optimization of the GPU-based CNN simulator is required, but GPU optimization is rarely straightforward with the current technology. Incidentally, the implementation presented here exhibits similar run-times as another GPU-based CNN simulator, which claims to be optimized for the hardware [6]. Clearly, further research is required to find effective optimization techniques applicable to CNN simulation.…”

Section: Y Conclusionmentioning

confidence: 91%

See 1 more Smart Citation

GPU-based simulation of cellular neural networks for image processing

Dolan

DeSouza

2009

2009 International Joint Conference on Neural Networks

View full text Add to dashboard Cite

improvement translates to faster image processing algorithms compared to traditional CPU-based algorithms. topology uniform 2D grid usually feed-forward processing element dynamic equations nonlinear weighted sum common uses image processing classification, control B. CNN TopologyCNNs are composed of many cells arranged in a grid, M.To simplify discussion, we will assume these grids are always square with dimensions m x m for m 2 cells. Each cell in the grid is denoted Vij for i, j EM. Thus each cell is labeled from VII to V m m .

show abstract

Section: Y Conclusionmentioning

confidence: 91%

“…Simple GPU-based CNN simulations have been demonstrated that run much faster than CPU-based CNN simulations [7], [6]. The research presented here examines whether this Abstract-The inherent massive parallelism of cellular neural networks makes them an ideal computational platform for kernelbased algorithms and image processing.…”

Section: Anns Cnnsmentioning

confidence: 99%

GPU-based simulation of cellular neural networks for image processing

Dolan

DeSouza

2009

2009 International Joint Conference on Neural Networks

View full text Add to dashboard Cite

show abstract

“…Simple GPU-based CNN simulations have already been demonstrated to run much faster than CPU-based CNN simulations [12,14]. Our goal here is to demonstrate that this improvement can translate into faster and easier image processing algorithms compared to traditional CPU-based algorithms.…”

Section: Introductionmentioning

confidence: 94%

CNN-based language and interpreter for image processing on GPUs

Dolan

DeSouza

2011

International Journal of Parallel, Emergent and Distributed Sys

View full text Add to dashboard Cite

The inherent massive parallelism of cellular neural networks (CNNs) makes them an ideal computational platform for kernel-based algorithms and image processing. General-purpose graphics processing units (GPUs) provide similar massive parallelism, but it can be difficult to design algorithms to make optimal use of the hardware. In this paper, we present a programming environment based on CNNs that can run on GPUs, multi-core CPUs or simply simulated in software. The platform offers a simplified view of massively parallel computation which remains universal and reasonably efficient. An image processing library with visualisation software has been developed using the abstraction to showcase the flexibility and power of cellular computation on GPUs. A simple virtual machine and language is presented to manipulate images using the library for single-core, multi-core and GPU backends.

show abstract

“…In addition, due to the structure of CNN, 2 International Journal of Reconfigurable Computing each layer of calculation is independent of others, and the interlayered structure can be dealt with like a flow structure. Moreover, various modules on FPGA can be executed in parallel [15]. It is very suitable to accelerate CNN by mapping the flow structure on FPGA.…”

Section: Introductionmentioning

confidence: 99%

Design of FPGA-Based Accelerator for Convolutional Neural Network under Heterogeneous Computing Framework with OpenCL

Luo

Qiao

et al. 2018

International Journal of Reconfigurable Computing

View full text Add to dashboard Cite

CPU has insufficient resources to satisfy the efficient computation of the convolution neural network (CNN), especially for embedded applications. Therefore, heterogeneous computing platforms are widely used to accelerate CNN tasks, such as GPU, FPGA, and ASIC. Among these, FPGA can accelerate the computation by mapping the algorithm to the parallel hardware instead of CPU, which cannot fully exploit the parallelism. By fully using the parallelism of the neural network's structure, FPGA can reduce the computing costs and increase the computing speed. However, the development of FPGA requires great design skills. As a heterogeneous development platform, OpenCL has some advantages such as high abstraction level, short development cycle, and strong portability, which can make up for the lack of skilled designers. This paper uses Xilinx SDAccel to realize the parallel acceleration of CNN task, and it also proposes an optimizing strategy of single convolutional layer to accelerate CNN. Simulation results show that the calculation speed could be improved by adopting the proposed optimizing strategy. Compared with the baseline design, the strategy of single convolutional layer could increase the computing speed 14 times. Performance of the whole CNN task could be improved 2 times more than before, and the speed of image classification could attain more than 48 fps.

show abstract

Cellular Neural Networks simulation on a parallel graphics processing unit

Cited by 12 publications

References 8 publications

GPU-based simulation of cellular neural networks for image processing

GPU-based simulation of cellular neural networks for image processing

CNN-based language and interpreter for image processing on GPUs

Design of FPGA-Based Accelerator for Convolutional Neural Network under Heterogeneous Computing Framework with OpenCL

Contact Info

Product

Resources

About