X-ray security inspection processes have a low degree of automation, long detection times, and are subject to misjudgment due to occlusion. To address these problems, this paper proposes a multi-objective intelligent recognition method for X-ray images based on the YOLO deep learning network and an optimized transformer structure (YOLO-T). We also construct the GDXray-Expanded X-ray detection dataset, which contains multiple types of dangerous goods. Using this dataset, we evaluated several versions of the YOLO deep learning network model and compared the results to those of the proposed YOLO-T model. The proposed YOLO-T method demonstrated higher accuracy for multitarget and hidden-target detection tasks. On the GDXray-Expanded dataset, the maximum mAP of the proposed YOLO-T model was 97.73%, which is 7.66%, 16.47%, and 7.11% higher than that obtained by the YOLO v2, YOLO v3, and YOLO v4 models, respectively. Thus, we believe that the proposed YOLO-T network has good application prospects in X-ray security inspection technologies. In all kinds of security detection scenarios using X-ray security detectors, the model proposed in this paper can quickly and accurately identify dangerous goods, which has broad application value.
High-resolution X-ray imaging diagnosis is a critical method for measuring Rayleigh-Taylor instability growth and hot spot interface morphology in inertial confinement fusion experiments. In this study, we develop a quasi-monochromatic elliptical Kirkpatrick–Baez microscope based on aberration theory, breaking the aberration limit of conventional Kirkpatrick–Baez microscopes. The microscope was characterized in the laboratory for spatial resolution performance and modulation transfer function before being implemented in cavity experiments at the SG-III prototype laser facility. The results demonstrate that the edge-based method achieves a spatial resolution of 20%.
The measurement of hotspot electron temperature is a paramount technique of implosion physics research in inertial confinement fusion. This study proposes a novel quasi-coaxis dual-energy flat spectral response high-resolution X-ray imaging instrument comprising a dual-channel total-reflection Kirkpatrick-Baez microscope and two flat non-periodic multilayer mirrors, which can image at 6.4 ± 0.5 and 9.67 ± 0.5 keV simultaneously. Various theoretical simulations were performed to verify the performance and feasibility of the imaging instrument, which was assembled and characterized in a laboratory. Experimental results show that the imaging instrument could achieve a high spatial resolution of 5 µm in a ± 150 µm field of view (FOV), the root mean square(RMS) deviation values of the measured reflection efficiency are 1.71% and 1.82% for the 6.4 keV and 9.67 keV imaging channels, respectively, in the ± 150 µm FOV.
1 October 2022 in Review of Scientific Instruments
Here, we describe a flat-field-response, four-channel x-ray imaging instrument developed to study hotspot asymmetries in inertial-confinement fusion experiments. We discuss the details of its design and optical characterization, the diagnostic deployment of the device, and experiments with it. We achieved a spatial-response flatness better than ∼8.4% within a ±200 µm field of view (FOV), with a spatial resolution of ∼4 µm at the center of the FOV. We used the system to characterize the low-order asymmetry of the implosion hotspot, and we obtained improved results after adjustments to improve the irradiation symmetry. Due to the flat-field-response characteristic, the versatile instrument also has the potential to be applied to diagnostics for the hotspot electron temperature and the Rayleigh–Taylor instability.
The diagnostic demand of hot-spot intensity distribution in laser inertial confinement fusion experiment has motivated quantitative measurement experiment to give the response efficiency of grazing incidence system to improve the accuracy of the diagnosis. In our study, a full-aperture intensity response measurement method is proposed, which is based on pinhole-scanning field of view (FOV) to obtain the response efficiency of grazing incidence system. The method is highly repeatable and enables response efficiency measurements over the full FOV with acceptable accuracy. A silicon positive-intrinsic-negative detector with an energy response range of 1.5 to 30 keV is used to directly detect the input and output spectrum. The response efficiency of a single-channel aspheric Kirkpatrick–Baez (KB) microscope coated with W/Si is obtained under the FOV of ±150 μm. The response efficiency of center FOV of the KB system is measured to be 2.52E-7 sr. The value is promising, as it shows a 6.6% maximum deviation from the theoretical value at center FOV, with a global uncertainty of 7.4%.
The quantitative measurement of plasma soft x-ray spectra is an important diagnostic problem in indirect-drive laser inertial confinement fusion (ICF). We designed, built, and tested a compact multichannel soft x-ray spectrometer with both spatial and temporal resolution capabilities for the detection of the spatiotemporal distribution of soft x-ray spectra. The spectrometer occupies a small solid angle, and the close measurement angle used for each channel enables the measurement of the angular distribution of emitting soft x-rays in ICF experiments. The spectrometer comprises pinhole, filter, and multilayer flat mirror arrays, and an x-ray streak camera. Its energy range is 0.1 – 3 keV. The dispersive elements of the spectrometer were calibrated at the Beijing Synchrotron Radiation Facility. The accuracy of the calibration was ≤ 5%, and the combined energy resolution (E/ΔE) of the calibrated dispersive elements of each channel was higher than 10. Finally, the instrument was tested at the Shenguang-III Laser Facility. The measurement results of x-ray radiation flux are agreed well with the experimental results of the M-band flat-response x-ray diode, demonstrating the feasibility of the proposed spectrometer configuration.