Aiming at the problem of blurred edge and low contrast of infrared images, this paper studies the improved median filter and improved Sobel edge detection to process infrared images. Based on the analysis of the features of the processed images, the improved Laplace pyramid decomposition image fusion algorithm is studied, and the purpose of rapid infrared image enhancement is achieved on a programmable GPU based on CUDA parallel processing technology. The algorithm combines the memory characteristics of GPU, applies texture mapping, multi-point access, and parallel triggering technology to optimize the data storage structure and improve the data processing speed. It is suitable for fields with high real-time requirements for infrared image enhancement. Experimental results show that the algorithm has good parallel characteristics, can make full use of the parallel computing capabilities of CUDA, improve the real-time performance of infrared image enhancement, and the speedup ratio reaches 32.189 when processing infrared images with a resolution of 3 096×3 096.
You Might Like
Recommended ContentMore
Open source project More
Popular Components
Searched by Users
Just Take a LookMore
Trending Downloads
Trending ArticlesMore