FAST TREE-STRUCTURAL IMAGE PROCESSING USING GPU(International Workshop on Advanced Image Technology 2007)

概要

論文の詳細を見る
Evolutionary Computation is an optimizing algorithm inspired by evolutional processes of living things. Recently, the researches of Evolutionary Computation such as Genetic Programming (GP) to image processing are popular. For example, the authors have already proposed Automatic Construction of Tree-structural Image Transformation (ACTIT). It automatically constructs an image-processing filter with GP. We have ever proven that ACTIT is an effective method for many problems. However, ACTIT requires much computing time in case we apply it to the problem which uses large and many training image sets. Therefore, we employ Graphics Processing Unit (GPU) to image processing for the purpose of reduction of time in this paper. Recently, the computational power of GPU is rapidly improving and outperforms that of CPU. The latest GPU has graphics pipelines which we can freely program and replace to perform complex CG processing. The research which puts GPU to practical use for the general purpose of calculating, General Purpose GPU (GPGPU) becomes popular. The processing speed of the proposed method is about 100 times faster than the ordinary ACTIT. We experimentally show that the proposed method is effective.
社団法人電子情報通信学会の論文
2007-01-02