Speeding up Mutual Information Computation Using NVIDIA CUDA Hardware
Speeding up Mutual Information Computation Using NVIDIA CUDA Hardware
复制标题
DOI：
10.1109/dicta.2007.4426846
复制
发表时间：
2007-12
期刊：
影响因子：
--
通讯作者：
R. Shams;Nick Barnes
中科院分区：
文献类型：
--
作者：
R. Shams;Nick Barnes
提出了一种用于NVIDIA‘计算统一设备架构’(CUDA)兼容设备的图像(2D或3D)之间互信息(MI)计算的有效方法。MI的高效并行化在图形处理器单元(GPU)上尤其具有挑战性，因为需要基于直方图计算具有大量箱的联合概率质量函数和边缘概率质量函数(PMF)。直方图更新的数据依赖(不可预测)性质，加上GPU的硬件限制(缺乏同步原语和有限的内存缓存机制)，可能会使基于GPU的计算效率低下。为了克服这些限制，我们使用联合直方图的下采样版本来近似PMF，从而避免了内存更新问题。与标准的基于CPU的实现相比，我们的CUDA实现将MI计算的效率提高了25倍，并且可以用于基于MI的图像配准应用。
We present an efficient method for mutual information (MI) computation between images (2D or 3D) for NVIDIA's `compute unified device architecture' (CUDA) compatible devices. Efficient parallelization of MI is particularly challenging on a `graphics processor unit' (GPU) due to the need for histogram-based calculation of joint and marginal probability mass functions (pmfs) with large number of bins. The data-dependent (unpredictable) nature of the updates to the histogram, together with hardware limitations of the GPU (lack of synchronization primitives and limited memory caching mechanisms) can make GPU-based computation inefficient. To overcome these limitation, we approximate the pmfs, using a down-sampled version of the joint- histogram which avoids memory update problems. Our CUDA implementation improves the efficiency of MI calculations by a factor of 25 compared to a standard CPU- based implementation and can be used in MI-based image registration applications.