Limits...
A Weighted Spatial-Spectral Kernel RX Algorithm and Efficient Implementation on GPUs

View Article: PubMed Central - PubMed

ABSTRACT

The kernel RX (KRX) detector proposed by Kwon and Nasrabadi exploits a kernel function to obtain a better detection performance. However, it still has two limits that can be improved. On the one hand, reasonable integration of spatial-spectral information can be used to further improve its detection accuracy. On the other hand, parallel computing can be used to reduce the processing time in available KRX detectors. Accordingly, this paper presents a novel weighted spatial-spectral kernel RX (WSSKRX) detector and its parallel implementation on graphics processing units (GPUs). The WSSKRX utilizes the spatial neighborhood resources to reconstruct the testing pixels by introducing a spectral factor and a spatial window, thereby effectively reducing the interference of background noise. Then, the kernel function is redesigned as a mapping trick in a KRX detector to implement the anomaly detection. In addition, a powerful architecture based on the GPU technique is designed to accelerate WSSKRX. To substantiate the performance of the proposed algorithm, both synthetic and real data are conducted for experiments.

No MeSH data available.


The architecture and operation of NVIDIA GPU, and the data transfer between CPU and GPU.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC5375727&req=5

sensors-17-00441-f005: The architecture and operation of NVIDIA GPU, and the data transfer between CPU and GPU.

Mentions: In Figure 5, “Host” represents the computer and “Device” represents GPUs. In device, there are different levels of memory. Thread private data will be assigned to the so-called local memory, when an excess of registers are used or the registers are depleted. In the computation, we would rather take full advantage of shared memory, which can be shared to threads in the same block to write/read quickly. The Global memory which could provide a wide memory bandwidth is supplied via Graphic Double Data Rate (GDDR) on the graphics card. It is a high-performance version of Double Data Rate (DDR) memory.


A Weighted Spatial-Spectral Kernel RX Algorithm and Efficient Implementation on GPUs
The architecture and operation of NVIDIA GPU, and the data transfer between CPU and GPU.
© Copyright Policy - open-access
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC5375727&req=5

sensors-17-00441-f005: The architecture and operation of NVIDIA GPU, and the data transfer between CPU and GPU.
Mentions: In Figure 5, “Host” represents the computer and “Device” represents GPUs. In device, there are different levels of memory. Thread private data will be assigned to the so-called local memory, when an excess of registers are used or the registers are depleted. In the computation, we would rather take full advantage of shared memory, which can be shared to threads in the same block to write/read quickly. The Global memory which could provide a wide memory bandwidth is supplied via Graphic Double Data Rate (GDDR) on the graphics card. It is a high-performance version of Double Data Rate (DDR) memory.

View Article: PubMed Central - PubMed

ABSTRACT

The kernel RX (KRX) detector proposed by Kwon and Nasrabadi exploits a kernel function to obtain a better detection performance. However, it still has two limits that can be improved. On the one hand, reasonable integration of spatial-spectral information can be used to further improve its detection accuracy. On the other hand, parallel computing can be used to reduce the processing time in available KRX detectors. Accordingly, this paper presents a novel weighted spatial-spectral kernel RX (WSSKRX) detector and its parallel implementation on graphics processing units (GPUs). The WSSKRX utilizes the spatial neighborhood resources to reconstruct the testing pixels by introducing a spectral factor and a spatial window, thereby effectively reducing the interference of background noise. Then, the kernel function is redesigned as a mapping trick in a KRX detector to implement the anomaly detection. In addition, a powerful architecture based on the GPU technique is designed to accelerate WSSKRX. To substantiate the performance of the proposed algorithm, both synthetic and real data are conducted for experiments.

No MeSH data available.