GPU-accelerated 2D OTSU and 2D entropy-based thresholding
GPU-accelerated 2D OTSU and 2D entropy-based thresholding
复制标题
GPU 加速的 2D OTSU 和基于 2D 熵的阈值处理
DOI：
10.1007/s11554-018-00848-5
复制
发表时间：
2019
影响因子：
3
通讯作者：
Yan Zheng
中科院分区：
文献类型：
--
作者：
Xianyi Zhu;Yi Xiao;Guanghua Tan;Shizhe Zhou;Andrew Chi-Sing Leung;Yan Zheng
图像阈值方法通常用于区分前景和背景方法。从$$ o（\ ell ^2）$$ o（ℓ2）到$$ o（\ ell ^4）$$ o（ℓ4），其中$$ \ ell $$是灰度的数量。本文提出了平行的算法（$$ o（\ ell + \ ell \ log \ ell）$$ O（ℓ +ℓlogℓ））通过将阈值分为七个级联的可平行方法，以加速2D OTSU和2D熵的阈值。计算步骤，我们的算法在GPU上执行所有计算，并且不需要GPU内存和主内存之间的数据传输。顺序算法（o（$$ \ ell ^2）$$ 2）。
Image thresholding methods are commonly used to distinguish foreground objects from a background. 2D thresholding methods consider both the value of a pixel and the mean of the pixel’s neighbors, so they are less sensitive to noises than 1D thresholding methods. However, the time complexity increases from $$O(\ell ^2)$$ O ( ℓ 2 ) to $$O(\ell ^4)$$ O ( ℓ 4 ) , where $$\ell$$ ℓ is the number of gray levels. This paper proposes a parallel algorithm ( $$O(\ell + \ell \log \ell )$$ O ( ℓ + ℓ log ℓ ) ) to accelerate both 2D OTSU and 2D entropy-based thresholding on GPU. By dividing the thresholding methods into seven cascaded parallelizable computational steps, our algorithm performs all the computations on GPU and requires no data transfer between GPU memory and main memory. The time complexity analysis explains the theoretical superiority over the state-of-the-art CPU sequential algorithm ( O ( $$\ell ^2)$$ ℓ 2 ) ). Experimental results show that our parallel thresholding runs 50 times faster than the sequential one without loss of accuracy.