计算机应用研究
計算機應用研究
계산궤응용연구
APPLICATION RESEARCH OF COMPUTERS
2013年
10期
3181-3184
,共4页
图像修补%整体变分%曲率驱动扩散%非线性各向异性扩散%P-Laplace算子%格子玻尔兹曼法
圖像脩補%整體變分%麯率驅動擴散%非線性各嚮異性擴散%P-Laplace算子%格子玻爾玆曼法
도상수보%정체변분%곡솔구동확산%비선성각향이성확산%P-Laplace산자%격자파이자만법
image inpaiting%total variation%curvature driven diffusions(CDD)%nonlinear anisotropic diffusion%P-Laplace operator%Lattice-Boltzmann method(LBM)
针对格子波尔兹曼(LB)模型模拟图像修复过程, 提出了一种基于P-Laplace算子的改进LB图像修复算法。该方法利用P-Laplace算子的非线性各向异性扩散性来填补受损区域, 并且忽略了梯度对修复过程的影响。实验结果表明, 改进后的LB修复模型, 其修复效果比TV和CDD模型好; 与LB模型具有相似的修复效果, 却比它具有更快的修复速率, 适用于苹果采摘机器人视觉系统中对树枝遮挡部分的修复、老照片修复, 以及文字移除等领域。
針對格子波爾玆曼(LB)模型模擬圖像脩複過程, 提齣瞭一種基于P-Laplace算子的改進LB圖像脩複算法。該方法利用P-Laplace算子的非線性各嚮異性擴散性來填補受損區域, 併且忽略瞭梯度對脩複過程的影響。實驗結果錶明, 改進後的LB脩複模型, 其脩複效果比TV和CDD模型好; 與LB模型具有相似的脩複效果, 卻比它具有更快的脩複速率, 適用于蘋果採摘機器人視覺繫統中對樹枝遮擋部分的脩複、老照片脩複, 以及文字移除等領域。
침대격자파이자만(LB)모형모의도상수복과정, 제출료일충기우P-Laplace산자적개진LB도상수복산법。해방법이용P-Laplace산자적비선성각향이성확산성래전보수손구역, 병차홀략료제도대수복과정적영향。실험결과표명, 개진후적LB수복모형, 기수복효과비TV화CDD모형호; 여LB모형구유상사적수복효과, 각비타구유경쾌적수복속솔, 괄용우평과채적궤기인시각계통중대수지차당부분적수복、로조편수복, 이급문자이제등영역。
To study Lattice-Boltzmann method(LBM) based image inpainting, this paper proposed an image inpainting algorithm based on P-Laplace operator. This algorithm made use of the nonlinear anisotropic diffusion of P-Laplace operator to inpaint damaged regions, and what's more, it ignored the effect caused by the gradient. Experimental results show that the proposed algorithm has better general performance in convergence speed and inpainting quality. Its objective is to restore the lost information according to around image information, which can be used to restore the shaded part of the branches in image processing system for apple harvesting robot, restore old photo, remove text and so on.
