Prompt gamma-ray activation image (PGAI) is a non-destructive element imaging method for large volume samples. Most of PGAI platforms are located in research reactors, which limit their applications. From the perspective of in-field applications, attractive alternative neutron sources are isotope neutron source and neutron generator. However, the neutron fluxes of these sources are much lower than that of reactor neutron source, which leads a poor spatial resolution.

This study aims to solve this problem by implementing an approach based on multi coded-aperture collimators.

First of all, the Monte Carlo code MCNP5 was employed to calculate spatial distribution of Cl in a known sample, and the characteristic gamma rays were produced by the thermal neutrons absorbed by the sample. Then, 36 coded-aperture collimators with random holes were used to collimate gamma rays, and 36 gamma signals were collected by high-purity germanium detectors (HPGe). Finally, the imaging of Cl was reconstructed through these data and maximum likelihood expectation maximization (MLEM) algorithm, and the relative deviation (d_f) and structural similarity (SSIM) were chosen to evaluate the image quality.

The spatial resolution of the imaging is 1 cm×1 cm, and the relative deviation and SSIM between the reconstructed image and the original image are 0.065 8 and 0.952 1, respectively. After neutron self-shielding correction, the relative deviation and SSIM between the reconstructed image and the original image are 0.002 3 and 0.998 4, respectively, which shows a good agreement.

The proposed approach is efficient to measure the distribution of Cl element, hence for element imaging of plate samples, and the reconstructed image is consistent with the set sample image.