CT image reconstruction back projection acceleration method based on OpenCL-To-FPGA

Abstract

The invention discloses a CT image reconstruction back projection acceleration method based on OpenCL-To-FPGA. Acceleration of the CT image reconstruction back projection step is achieved through an FPGA. The method comprises the specific steps that a CPU-FPGA heterogeneous computing mode with a CPU and the FPGA cooperating with each other is constructed in an OpenCL programming model, the CPU and the FPGA are in communication through a PCI-E bus, the CPU serves as a host and is in charge of serial tasks in an algorithm and the tasks of configuration and control on the FPGA, and the FPGA serves as a coprocessor and achieves parallel pipeline acceleration of back projection computing by loading an OpenCL kernel program. In the programming mode, executive programs of the FPGA are all developed through an OpenCL language similar to C / C++ in style, development is easy and convenient to perform, modification is flexible, the development cycle can be greatly shortened, and the development cost for product maintenance and upgrading is reduced; moreover, the new method is based on an OpenCL frame, codes can be fast transplanted between platforms, and the method is suitable for being extended and applied to cooperative acceleration of a multi-processor heterogeneous platform.

Description

technical field [0001] The invention relates to a method for CT image reconstruction, in particular to a back-projection acceleration method in the CT image reconstruction process based on OpenCL-To-FPGA. Background technique [0002] Computed Tomography (CT) has been widely used in the fields of industrial non-destructive testing and medical diagnosis due to its advantages of non-destructive, accurate and three-dimensional visualization. However, while CT equipment improves the imaging quality and reconstruction resolution, the corresponding computing resources and storage resources are very large. The reconstruction speed has become a bottleneck in the practical application of CT technology. How to accelerate the CT reconstruction process is currently the research field in this field. A difficult question. [0003] At present, the mainstream algorithms used in actual CT systems mainly focus on analytical reconstruction algorithms based on filtered back projection, such as...

AI Technical Summary

Problems solved by technology

[0005] Through the literature search of the prior art, it is found that the implementation method adopted in the article of using FPGA to accelerate CT reconstruction back projection has a long development cycle, is difficult to quickly transplant across platforms, and is not suitable for Multi-processor heterogeneous platform cooperative acceleration and other deficiencies

In 2008, Markus Kowarschik et al. published the article "High-Speed ​​Reconstruction for C-Arm Computed Tomography" on Proceedings of the 9th International Meeting on Fully Three-Dimensional Image Reconstruction in Radiology and Nuclear Medicine, using 9 FPGAs to coordinate and complete CT The acceleration of the 3D back-projection part in the reconstruction, but the development method he adopted is based on the traditional VHDL / Verilog hardware programming language, the development complexity is high, the development cycle is long, and the development program can only be used on the FPGA and cannot be transplanted to other processing device

[0006]In 2009, in his dissertation "An FPGA Hardware Solution for Accelerating Tomographic Reconstruction", Xu proposed to use the Impulse C language to implement the FBP algorithm for CT reconstruction on FPGA. The FPGA accelerated the two-dimensional back-projection process in the FBP algorithm. The Impulse C language adopts a C-like programming mode, and the development speed and modification flexibility are stronger than the traditional development methods. However, Impulse C is only developed for FPGA programming and still exists. Difficult to transplant, not suitable for multi-processor heterogeneous platform collaborative acceleration and other shortcomings

Images