基于FPGA+GPU的图像采集处理系统设计

Design of image acquisition and processing system based on FPGA+GPU

摘要:

随着嵌入式图像处理系统的快速发展，对于前端图像采集模块的需求越来越高。图像采集的速度、分辨率、可靠性以及集成度对后续设计的准确度由极大的影响。通过对数字图像采集系统进行研究，设计出了基于FPGA和GPU架构的图像采集处理系统，重点研究了图像采集处理系统的硬件设计过程和软件设计过程。在基于FPGA+GPU的图像采集处理系统中，让具有强大运算处理能力的GPU专注于数据存储、用户交互以及后续的图像处理。系统中，FPGA则负责图像的采集、外设控制、任务调度。GPU与FPGA之间通过高速PCIE总线进行通信，分别设计编写基于Linux系统的驱动程序和FPGA端PCIE程序。实验结果表明，所设计基于FPGA+GPU的图像采集处理系统可实现437.5Mbps的实时图像采集存储速度，传输过程实时稳定，数据传输完整。

Abstract:

With the rapid development of embedded image processing system, the demand for front-end image acquisition modules is increasing. The speed, resolution, reliability, and integration of image acquisition greatly affect the accuracy of subsequent image processing. By studying the digital image acquisition system, an image acquisition and processing system based on FPGA and GPU architecture was designed, with focusing on the hardware and software design process of the image acquisition and processing system. In the image acquisition and processing system based on FPGA+GPU, the GPU with powerful computing and processing capabilities is focused on data storage, user interaction, and subsequent image processing. In the system, FPGA is responsible for image acquisition, peripheral control, and task scheduling. The communication between GPU and FPGA is carried out through high-speed PCIE bus, and driver programs based on Linux system and FPGA PCIE programs are designed separately. The experimental results show that the designed image acquisition and processing system based on FPGA+GPU can achieve a real-time image acquisition and storage speed of 437.5Mbps. The transmission process is real-time and stable, and the transmitted data is complete.