• Journal of Infrared and Millimeter Waves
  • Vol. 40, Issue 1, 108 (2021)
Wen-Jun SHI1, Deng-Wei WANG2、3、*, Wan-Suo LIU1, and Da-Gang JIANG2、3
Author Affiliations
  • 1Aviation Maintenance School for NCO,Air Force Engineering University,Xinyang 464000,China
  • 2School of Aeronautics and Astronautics,University of Electronic Science and Technology of China,Chengdu 611731,China
  • 3Aircraft Swarm Intelligent Sensing and Cooperative Control Key Laboratory of Sichuan Province,Chengdu 611731,China
  • show less
    DOI: 10.11972/j.issn.1001-9014.2021.01.016 Cite this Article
    Wen-Jun SHI, Deng-Wei WANG, Wan-Suo LIU, Da-Gang JIANG. GPU accelerated level set model solving by lattice boltzmann method with application to image segmentation[J]. Journal of Infrared and Millimeter Waves, 2021, 40(1): 108 Copy Citation Text show less

    Abstract

    A novel Graphics Processing Units (GPU) accelerated level set model which organically combines the global fitting energy and the local fitting energy from different models and the weighting coefficient of the global fitting term can be adaptively adjusted, is proposed to image segmentation. The proposed model can efficiently segment images with intensity inhomogeneity regardless of where the initial contour lies in the image. In its numerical implementation, an efficient numerical scheme called Lattice Boltzmann Method (LBM) is used to break the restrictions on time step. In addition, the proposed LBM is implemented by using a NVIDIA GPU to fully utilize the characteristics of LBM method with high parallelism. The extensive and promising experimental results from synthetic and real images demonstrate the effectiveness and efficiency of the proposed method.In addition, the factors that can have a key impact on segmentation performance are also analyzed in depth.
    ECVϕ=λ1CVΩIx-Min2Hϕxdx+               λ2CVΩIx-Mout21-Hϕxdx+               μΩδϕxϕxdx+νΩHϕxdx ,

    View in Article

    ϕt=δϕμdivϕϕ-ν-λ1CVI-Min2+λ2CVI-Mout2 ,

    View in Article

    Min(ϕ)=ΩI(x)H(ϕ(x))dxΩH(ϕ(x))dxdyMout(ϕ)=ΩI(x)(1-H(ϕ(x)))dxΩ(1-H(ϕ(x)))dx ,

    View in Article

    Ex=λ1LBFinsideCKσx-yIy-f1x2dydx+        λ2LBFoutsideCKσx-yIy-f2x2dydx,

    View in Article

    ELBF=ΩExϕ,f1x,f2xdx       =λ1LBFΩΩKσx-yIy-f1x2Hϕydydx+λ2LBFΩΩKσx-yIy-f2x21-Hϕydydx ,

    View in Article

    ϕt=-δϕλ1LBFe1-λ2LBFe2 ,

    View in Article

    e1x=ΩKσy-xIx-f1y2dye2x=ΩKσy-xIx-f2y2dy ,

    View in Article

    f1x=Kσ*HεϕIxKσ*Hεϕf2x=Kσ*1-HεϕIxKσ*1-Hεϕ .

    View in Article

    Hεz=121+2πarctanzεδεz=1πεε2+z2,  zR .

    View in Article

    Elocalϕ=λ1CVΩΩKσx-yIy-f1x2Hϕydydx+λ2CVΩΩKσx-yIy-f2x21-Hϕydydx .

    View in Article

    Eglobalϕ=λ1LBFΩIx-Min2Hϕxdx+λ2LBFΩIx-Mout21-Hϕxdx .

    View in Article

    Eϕ=Elocalϕ+wEglobalϕ ,

    View in Article

    LCRN=Vmax-VminL ,

    View in Article

    w=κmeanLCRN.1-LCRN ,

    View in Article

    Pϕ=12ϕx-12dx .

    View in Article

    Lϕ=Hϕxdx .

    View in Article

    Eour=E+μPϕ+νLϕ ,

    View in Article

    ϕt=δεϕ-λ1CVI-Min2+λ2CVI-Mout2- wδεϕλ1LBFe1-λ2LBFe2+ μΔϕ-divϕϕ+νδεϕdivϕϕ ,

    View in Article

    fir+eiΔt,t+Δt=fir,t+1τfieqr,t-fir,t+Dbc2Fei ,

    View in Article

    fieqρ,u=ρAi+Bieiu+Cieiu+Diu2 ,

    View in Article

    ρ=ifiu=1ρifiei ,

    View in Article

    fieqρ,u=ρAi .

    View in Article

    fieq=ραi1+3eiu+92eiu2-32u2, i=0,1,,8 ,

    View in Article

    γ=292τ-1 .

    View in Article

    ρt=γ ρ+F .

    View in Article

    ρt=γ ρ+F=γdivϕϕ+F ,

    View in Article

    fir+eiΔt,t+Δt=fir,t+1τfieqr,t-fir,t+Dbc2wδεϕ-λ1I-Min2+λ2I-Mout2-δεϕI-ILIFu1-u2+μΔϕ .

    View in Article

    1imageGPU=gpuArrayimage;2ModelParameterStructureGPU=gpuArrayModelParameterStructure;3FGPU=arrayfun@body_force,imageGPU,ModelParameterStructure;4F=gatherFGPU.(28)

    View in Article

    JS=NSreferenceStestNSreferenceStest ,

    View in Article

    DSC=2NSreferenceStestNSreference+NStest ,

    View in Article

    ϕinitialx,y=-c, x,yΩinitial-Ωinitial 0,   x,yΩinitialc,   x,yΩ-Ωinitial ,

    View in Article

    Wen-Jun SHI, Deng-Wei WANG, Wan-Suo LIU, Da-Gang JIANG. GPU accelerated level set model solving by lattice boltzmann method with application to image segmentation[J]. Journal of Infrared and Millimeter Waves, 2021, 40(1): 108
    Download Citation