Speedup segmentation of industrial DR image by CUDA
NI Feng-yuea · Jisuanji yingyong yanjiu · 2011
The segmentation of industrial DR image is very important to industrial image processing.C-V method has some advantage to industrial DR image segmentation.Due to its higher computing complexity,C-V method is rarely applied in a time-critical industrial environment.This paper presented an efficient parallel implementation of DR image segmentation by C-V method with CUDA,and used a method that separated data computing merged in corresponding thread by shared memory.Finally,compared the results to the current applications on image processing workstation,the implementation exhibited a speed-up of 32 to 44 times compared to a state-of-the-art CPU.