GPU Implementation of Spiking Neural Networks for Color Image Segmentation

Ermai Xie, TM McGinnity, Qingxiang Wu, Jianyong Cai, Rongtai Cai

    Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

    5 Citations (Scopus)

    Abstract

    Spiking neural networks (SNN) are powerful computational model inspired by the human neural system for engineers and neuroscientists to simulate intelligent computation of the brain. Inspired by the visual system, various spiking neural network models have been used to process visual images. However, it is time-consuming to simulate a large scale of spiking neurons in the networks using CPU programming. Spiking neural networks inherit intrinsically parallel mechanism from biological system. A massively parallel implementation technology is required to simulate them. To address this issue, modern Graphic Processing Units (GPUs), which have parallel array of streaming multiprocessors, allow many thousands of lightweight threads to be run, is proposed and proved as a pertinent solution. This paper presents an approach for implementation of an SNN model which performs color image segmentation on GPU. This approach is then compared with an equivalent implementation on an Intel Xeon CPU. The results show that the GPU approach was found to provide a 31 times faster than the CPU implementation.
    LanguageEnglish
    Title of host publicationUnknown Host Publication
    Place of PublicationUSA
    Pages1260-1264
    Number of pages5
    Publication statusPublished - 10 Oct 2011
    Event4th International Congress on Image and Signal Processing (CISP 2011) -
    Duration: 10 Oct 2011 → …

    Conference

    Conference4th International Congress on Image and Signal Processing (CISP 2011)
    Period10/10/11 → …

    Fingerprint

    Image segmentation
    Color
    Neural networks
    Program processors
    Biological systems
    Computer programming
    Neurons
    Brain
    Engineers
    Graphics processing unit

    Cite this

    Xie, E., McGinnity, TM., Wu, Q., Cai, J., & Cai, R. (2011). GPU Implementation of Spiking Neural Networks for Color Image Segmentation. In Unknown Host Publication (pp. 1260-1264). USA.
    Xie, Ermai ; McGinnity, TM ; Wu, Qingxiang ; Cai, Jianyong ; Cai, Rongtai. / GPU Implementation of Spiking Neural Networks for Color Image Segmentation. Unknown Host Publication. USA, 2011. pp. 1260-1264
    @inproceedings{85bf2d02490346359fea55b705370065,
    title = "GPU Implementation of Spiking Neural Networks for Color Image Segmentation",
    abstract = "Spiking neural networks (SNN) are powerful computational model inspired by the human neural system for engineers and neuroscientists to simulate intelligent computation of the brain. Inspired by the visual system, various spiking neural network models have been used to process visual images. However, it is time-consuming to simulate a large scale of spiking neurons in the networks using CPU programming. Spiking neural networks inherit intrinsically parallel mechanism from biological system. A massively parallel implementation technology is required to simulate them. To address this issue, modern Graphic Processing Units (GPUs), which have parallel array of streaming multiprocessors, allow many thousands of lightweight threads to be run, is proposed and proved as a pertinent solution. This paper presents an approach for implementation of an SNN model which performs color image segmentation on GPU. This approach is then compared with an equivalent implementation on an Intel Xeon CPU. The results show that the GPU approach was found to provide a 31 times faster than the CPU implementation.",
    author = "Ermai Xie and TM McGinnity and Qingxiang Wu and Jianyong Cai and Rongtai Cai",
    year = "2011",
    month = "10",
    day = "10",
    language = "English",
    isbn = "978-1-4244-9305-0",
    pages = "1260--1264",
    booktitle = "Unknown Host Publication",

    }

    Xie, E, McGinnity, TM, Wu, Q, Cai, J & Cai, R 2011, GPU Implementation of Spiking Neural Networks for Color Image Segmentation. in Unknown Host Publication. USA, pp. 1260-1264, 4th International Congress on Image and Signal Processing (CISP 2011), 10/10/11.

    GPU Implementation of Spiking Neural Networks for Color Image Segmentation. / Xie, Ermai; McGinnity, TM; Wu, Qingxiang; Cai, Jianyong; Cai, Rongtai.

    Unknown Host Publication. USA, 2011. p. 1260-1264.

    Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

    TY - GEN

    T1 - GPU Implementation of Spiking Neural Networks for Color Image Segmentation

    AU - Xie, Ermai

    AU - McGinnity, TM

    AU - Wu, Qingxiang

    AU - Cai, Jianyong

    AU - Cai, Rongtai

    PY - 2011/10/10

    Y1 - 2011/10/10

    N2 - Spiking neural networks (SNN) are powerful computational model inspired by the human neural system for engineers and neuroscientists to simulate intelligent computation of the brain. Inspired by the visual system, various spiking neural network models have been used to process visual images. However, it is time-consuming to simulate a large scale of spiking neurons in the networks using CPU programming. Spiking neural networks inherit intrinsically parallel mechanism from biological system. A massively parallel implementation technology is required to simulate them. To address this issue, modern Graphic Processing Units (GPUs), which have parallel array of streaming multiprocessors, allow many thousands of lightweight threads to be run, is proposed and proved as a pertinent solution. This paper presents an approach for implementation of an SNN model which performs color image segmentation on GPU. This approach is then compared with an equivalent implementation on an Intel Xeon CPU. The results show that the GPU approach was found to provide a 31 times faster than the CPU implementation.

    AB - Spiking neural networks (SNN) are powerful computational model inspired by the human neural system for engineers and neuroscientists to simulate intelligent computation of the brain. Inspired by the visual system, various spiking neural network models have been used to process visual images. However, it is time-consuming to simulate a large scale of spiking neurons in the networks using CPU programming. Spiking neural networks inherit intrinsically parallel mechanism from biological system. A massively parallel implementation technology is required to simulate them. To address this issue, modern Graphic Processing Units (GPUs), which have parallel array of streaming multiprocessors, allow many thousands of lightweight threads to be run, is proposed and proved as a pertinent solution. This paper presents an approach for implementation of an SNN model which performs color image segmentation on GPU. This approach is then compared with an equivalent implementation on an Intel Xeon CPU. The results show that the GPU approach was found to provide a 31 times faster than the CPU implementation.

    M3 - Conference contribution

    SN - 978-1-4244-9305-0

    SP - 1260

    EP - 1264

    BT - Unknown Host Publication

    CY - USA

    ER -

    Xie E, McGinnity TM, Wu Q, Cai J, Cai R. GPU Implementation of Spiking Neural Networks for Color Image Segmentation. In Unknown Host Publication. USA. 2011. p. 1260-1264