Authors' Contribution LZ conceived and designed the experiments. XW and EY performed the experiments, participated in the field work and analyzed the data. LZ, XW and EY wrote the manuscript.
In this paper, an accelerated JPEG(Joint Photographic Experts Group) decoder was efficiently implemented on the GPU(Graphic Processing Unit) using the CUDA(Computer Unified Device Architecture) technology and it is capable for high definition images decoding. The CUDA technology can assist the GPU to work for the CPU for large computation. In this paper, the IDCT(Inverse Discrete Cosine Transform) model which has consumed about 75% the total time in the JPEG decoder is worked in the GPU by the CUDA, and other models of the JPEG decoder are designed in the CPU. At the same time, we use the asynchronous parallel execution between the CPU and the GPU to improve the JPEG decoder acceleration rate. In the experiment, the JPEG decoder based on the CUDA performs decompression of 3240 x 2160 pixels images, the implementation of the CUDA-based IDCT can be more than 49 times faster than the implementation on the CPU and the total processing time of the whole JPEG decoder can save about 50% time than the CPU.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.