29 May 2024 in IEEE Transactions on Instrumentation and Measurement
The size and stability of the X-ray beam focal spot in the electron linear accelerator are crucial factors affecting the spatial resolution of high-energy industrial computed tomography. The commonly used “Sandwich” test method for the X-ray beam focal spot size measurement is complex, vulnerable to human-related factors, and impossible for real-time measurement of stability. In this paper, we devise a slit translation scanning method to measure both size and stability of the X-ray beam focal spot for the electron linear accelerator. Firstly, this paper theoretically analyzed the impact of the effective slit width and sampling frequency on the measurement results. Secondly, a 9 MeV electron linear accelerator was used to measure the X-ray beam focal spot size. The measurement errors along x and y directions are smaller than 5.6% compared to the results measured by the IEC 62976-2017 standard method. Last, The stability of the X-ray beam focal spot was assessed using the slit translation scanning method with three different electron linear accelerators (9 MeV, 6 MeV, and 2 MeV). The results reveal random drift of the X-ray beam focal spot position, with a maximum drift up to 0.6 mm. The experiments demonstrate the simplicity and efficiency of the proposed slit translation scanning method which proves invaluable for the meticulous measurement of the size and stability of the X-ray beam focal spot for electron linear accelerator.
26 March 2025 in IEEE Transactions on Instrumentation and Measurement
To achieve high spatial resolution of X-ray imaging, the focal spot size of high-energy non-destructive testing electron accelerators has been progressively reduced from 2 mm to below 1 mm, and in some cases, down to tens of micrometers. To measure these increasingly smaller focal spots, traditional methods such as the “Sandwich” test method and the slit translation scanning method, are no longer effective. This paper developed an improved circular edge analysis method (ICEAM) for measuring the size and shape of the focal spot of electron accelerators. The ICEAM employs the detector’s point spread function to correct the non-ideal circular aperture image caused by the X-ray scatter and the detector’s cross-talk. Then, edge spread functions of the corrected circular aperture image are used to construct a sinogram of the focal spot, which is subsequently used to reconstruct the intensity distribution image of the focal spot by using the filtered back-projection method. This approach enables precise measurement of both the shape and the size of the focal spot. Theoretical and experimental results demonstrate that ICEAM can not only conveniently and rapidly measure the shape and size of focal spots for conventional high-energy electron accelerators, with measurement errors within 5.8%, but also accurately measure the micrometer-scale focal spots for novel MeV energy electron accelerators (Rhodotron), with measurement errors within 16.2% of the theoretical value. The proposed ICEAM effectively tackles the challenges in measuring the shape and size of the microfocus for novel high-energy electron accelerators, and it is of significant importance for advancing the application of these electron accelerators in digital X-ray imaging and high-end equipment such as industrial CT.
5 August 2026 in SSRN Electronic Journal
Sparse-view computed tomography (CT) reduces radiation dose and acquisition time, but the limited number of projections makes the reconstruction problem highly underdetermined, leading to streak artifacts and structural degradation. Existing 3D Gaussian Splatting (3DGS)-based CT reconstruction methods typically initialize Gaussian primitives from full FDK point clouds, causing streak artifacts and beam-hardening responses in sparse-view FDK reconstructions to be inherited by the initial Gaussian distribution. Moreover, limited projection supervision tends to drive Gaussian primitives into fixed co-adaptation relationships, resulting in needle-like artifacts during optimization. To address these issues, we propose a sparse-view CT reconstruction method that combines Sparse-Point Initialization (SPI) with Decaying Opacity Perturbation (DOP). SPI selects representative structural points from the FDK reconstruction through joint density-gradient scoring and improves spatial coverage using Adaptive Boundary Expansion (ABE), thereby reducing artifact contamination in the initial Gaussian distribution. DOP applies iteration-decaying multiplicative Laplace noise to Gaussian opacities during training, continuously modulating Gaussian contributions to suppress fixed co-adaptation. Experiments on the public FIPS dataset and self-built industrial CT datasets demonstrate that the proposed method improves PSNR by up to 1.132~dB and SSIM by up to 0.023 over the R2-Gaussian baseline under 25-view reconstruction, while effectively reducing the Co-Adaptation Score (CA).