Fast Local Contrast Enhancement Using GPU Computing in Ultrasound Image
Yang Wang, Changyong Deng, Bo Peng · International Journal of Applied Mathematics & Statistics/International journal of applied mathematics and statistics · 2013
A graphics processing unit (GPU) computing approach to enhance the local contrast of ultrasound images based on multi-scale morphology is presented in this paper. The original enhancement method we proposed involves local histogram matching, top-hat transformation, and other computationally intensive processes. In this paper, we describe the use of a GPU to implement our proposed local contrast enhancement of ultrasound images. This approach can achieve superior computational efficiency for 512 (height) x 330 (width) images compared to CPU implementation, up to 156 times faster. Testing results using GPU and CPU demonstrate that the quality of images using parallel algorithms is the same, but at the same time, there is a significant improvement in computational efficiency.