FPGA Acceleration of a Supervised Learning Method for Hyperspectral Image Classification

Tajiri, Kento; Matsumoto, Tsutomu

doi:10.1109/fpt.2018.00051

Cited by 6 publications

(11 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…So far, we have achieved a maximum FPGA performance of 55GOPS, which is not bad compared to a similar solution [12][13] [16]. They are implementing larger image processing networks such as AlexNet [15][18], VGG, etc.…”

Section: Network Inferencementioning

confidence: 99%

Comparison of Performance of Optimized HSI CNN models on Desktop and Embedded Platforms

Čech

Rozkovec

2021

2021 International Conference on Applied Electronics (AE)

View full text Add to dashboard Cite

We compare different platforms for inference of convolutional neural networks in this paper. We trained various neural networks to determine the material in the source hyperspectral cube. Then we convert them to inference format and compare the inference results. We used tools under Xilinx Vitis AI for FPGA implementation. We try to minimize the size of the proposed networks by pruning them and provide further comparisons. FPGA platforms show to be energy efficient but still slower than a graphics card in terms of performance.

show abstract

Section: Network Inferencementioning

confidence: 99%

Comparison of Performance of Optimized HSI CNN models on Desktop and Embedded Platforms

Čech

Rozkovec

2021

2021 International Conference on Applied Electronics (AE)

View full text Add to dashboard Cite

show abstract

“…The accelerator uses data flow programming to achieve high performance and can be used for different applications. Tajiri et al [18] proposed a hyperspectral image classification system on FPGA, by introducing the Composite Kernel Support Vector Machine and reducing the computational complexity. These former accelerators achieve real time processing speed but they do not achieve high classification accuracy and therefore are not favoured over CNN-based methods.…”

Section: Related Workmentioning

confidence: 99%

“…It should be noted that CPUs and GPUs are not realistic to be mount on a satellite or a drone because of their high power consumption, and therefore their usability is very limited in space platforms. Finally we compare our accelerator to other FPGA-based accelerators implementing SVM [19,18]. These two accelerator are implemented in an Altera Stratix V 5SGSMD8N2F45C2 FPGA on Maxeler MAX4 DFE [19], and in a Xilinx Kintex-7 XC7K325T-FF2-900 FPGA device [18] respectively.…”

Section: Performance Comparison Vs Other Processors and Acceleratorsmentioning

confidence: 99%

“…Finally we compare our accelerator to other FPGA-based accelerators implementing SVM [19,18]. These two accelerator are implemented in an Altera Stratix V 5SGSMD8N2F45C2 FPGA on Maxeler MAX4 DFE [19], and in a Xilinx Kintex-7 XC7K325T-FF2-900 FPGA device [18] respectively. Due to these accelerators have different DSP numbers compared to ours, we also provide the speedups normalized by the number of DSPs.…”

Section: Performance Comparison Vs Other Processors and Acceleratorsmentioning

confidence: 99%

“…3.4x 1x * The numbers in this row represent the speedups of our model run in FPGA platform compared to that run in CPU and GPU platforms. .0 b a This is the power consumption reported in [18]. b The power consumption of ours is measured from the board using a power meter.…”

Section: Performance Comparison Vs Other Processors and Acceleratorsmentioning

confidence: 99%

See 2 more Smart Citations

Optimizing CNN-Based Hyperspectral Image Classification on FPGAs

Liu

Chu

Wang

et al. 2019

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Hyperspectral image (HSI) classification has been widely adopted in applications involving remote sensing imagery analysis which require high classification accuracy and real-time processing speed. Methods based on Convolutional neural networks (CNNs) have been proven to achieve state-of-the-art accuracy in classifying HSIs. However, CNN models are often too computationally intensive to achieve real-time response due to the high dimensional nature of HSI, compared to traditional methods such as Support Vector Machines (SVMs). Besides, previous CNN models used in HSI are not specially designed for efficient implementation on embedded devices such as FPGAs. This paper proposes a novel CNN-based algorithm for HSI classification which takes into account hardware efficiency. A customized architecture which enables the proposed algorithm to be mapped effectively onto FPGA resources is then proposed to support real-time on-board classification with low power consumption. Implementation results show that our proposed accelerator on a Xilinx Zynq 706 FPGA board achieves more than 70× faster than an Intel 8-core Xeon CPU and 3× faster than an NVIDIA GeForce 1080 GPU. Compared to previous SVM-based FPGA accelerators, we achieve comparable processing speed but provide a much higher classification accuracy.

show abstract

FPGA Acceleration of a Composite Kernel SVM for Hyperspectral Image Classification

Tajiri

Matsumoto

2023

IEEE Access

View full text Add to dashboard Cite

Hyperspectral image classification is one of the most important techniques for analyzing hyperspectral image that have hundreds of spectrum luminance values of near-infrared to visible light. For this classification, supervised learning methods are widely used, but in general, they typically trade off between their accuracy and computational complexity. Our approach is based on a composite kernel method, and the computation is simplified to achieve higher processing speeds by efficiently using the hardware resources of FPGA. The accuracy of this approach reaches 98.0% and 98.8% on two benchmark datasets, Indian Pines and Salinas via simulation, which is comparable to those in previous works. Two implementations, one with less hardware resources but more off-chip memory bandwidth, and another with more hardware resources but less off-chip memory bandwidth, are implemented on an FPGA and evaluated. The processing speeds of the two implementations are the same, which is 1.3 Mpixels/s for 2048 pixel wide images. This processing speed is fast enough for real-time processing, and faster than previous studies when normalized by hardware size and power consumption. We also introduce two more implementations that aim to reduce the on-chip memory usage of the second implementation within a reasonable increase of off-chip memory bandwidth, and we discuss which implementation is advantageous under what conditions.

show abstract

FPGA Acceleration of a Supervised Learning Method for Hyperspectral Image Classification

Cited by 6 publications

References 7 publications

Comparison of Performance of Optimized HSI CNN models on Desktop and Embedded Platforms

Comparison of Performance of Optimized HSI CNN models on Desktop and Embedded Platforms

Optimizing CNN-Based Hyperspectral Image Classification on FPGAs

FPGA Acceleration of a Composite Kernel SVM for Hyperspectral Image Classification

Contact Info

Product

Resources

About