Title: Design of real-time digital image processing system based on high performance computing
Authors: Hongxing Sun; Yingwei Zhang; Wei Teng
Addresses: School of Information Technology, Nanchang Vocational University, Nanchang, Jiangxi, China ' School of Electronic and Information Engineering, University of Science and Technology Liaoning, Anshan, Liaoning, China ' School of Computer Science and Software Engineering, University of Science and Technology Liaoning, Anshan, Liaoning, China
Abstract: Nowadays, digital image processing systems are developing towards faster data transmission and processing speed, smaller size, higher real-time performance and more flexible and convenient programming. How to improve the performance of digital image processing systems has become a concern for people. Starting from the actual project development needs, this article uses high-performance computing technology and combines the real-time processing characteristics of images. With TMS320C6455 as the core framework, Field Programmable Gate Array (FPGA) + Digital Signal Processing (DSP) technology is used to achieve high-performance computing and real-time image processing in digital image processing systems. This article uses TMS320C6455 embedded in FPGA to achieve data caching and data sharing between nodes. On this basis, real-time pre-processing of digital images is carried out using FPGA, and a high-speed serial interface with FPGA is used to increase the transmission bandwidth of the system. In the real-time tracking testing of images, the high-performance real-time digital image processing system used in this article took an average of 0.00425 s per frame of image and 0.00235 s in the process of feature extraction and matching. The high-performance real-time digital image processing system used in this article can achieve real-time image processing.
Keywords: real-time digital image processing system; high-performance computing; image features; digital image pre-processing.
DOI: 10.1504/IJGUC.2024.140120
International Journal of Grid and Utility Computing, 2024 Vol.15 No.3/4, pp.352 - 360
Received: 31 May 2023
Accepted: 08 Aug 2023
Published online: 24 Jul 2024 *