Parallel optical flow estimation by dividing image in section
Parallel optical flow estimation by dividing image in section
复制标题
通过分割图像进行并行光流估计
DOI：
10.1109/iccas.2015.7364935
复制
发表时间：
2015
期刊：
影响因子：
--
通讯作者：
Hiroshi Harada
中科院分区：
文献类型：
--
作者：
Yusuke Miyajima;Teruo Yamaguchi;Hiroshi Harada
时空微分法是确定运动目标速度的最有效方法之一。然而，光流估计需要很长的时间，从而难以实现实时目标跟踪。在这项研究中，我们评估了并行计算时的光流估计所消耗的时间。并行化的方法包括将输入图像划分为若干区域，并将每个区域分配给每个多核。由于时空微分法是由一个局部计算组成的，所以它与图像分割的并行化方法是兼容的。我们可以使用商业多核处理器来加速光流估计。实验结果表明，该方法对加速光流场估计是可行的。实验结果表明，将计算区域划分为相同核数的部分时，可以获得最佳的加速效果。
Spatio-temporal differentiation method is one of the most effective methods of determining the velocity of moving objects. However, the optical flow estimate takes long time so that real-time object tracking is difficult to realize. In this study, we evaluated consuming time of the optical flow estimation when calculating in parallel. The method of parallelization includes dividing input image into several region and assigning each region into each multi-core. Because spatio-temporal differentiation method is composed of a local calculation, it is compatible with the parallelization method of dividing image into section. We can use the commercial multi-core processors to accelerate the optical flow estimation. It was found feasible to accelerate the optical flow estimation by the method. Experimental result shows that the most suitable acceleration can be achieved when dividing calculation region into parts of the same number of the cores.