The measurement of the electron temperature distribution of hotspot requires bremsstrahlung imaging with both spectral resolution and spatial resolution. The Kirkpatrick–Baez (KB) microscope is an important diagnostic device in inertial confinement fusion experiments, which can realize high spatial resolution imaging of multi-energy. However, conventional KB microscopes face a fundamental trade-off between achieving a quasi-monoenergetic response and maintaining uniform efficiency across the field of view. This inherent limitation leads to difficulties in efficiency calibration and can introduce distortion in hotspot images, thereby preventing their effective application in the quantitative measurement of hotspot radiation intensity. This paper proposes a four-energy-point KB microscope with a self-adjusting spectral response, in which the response energies adapt to the field-of-view position. The design simultaneously achieves a quasi-monoenergetic response and a uniform efficiency distribution, thereby mitigating hot-spot image distortion and simplifying efficiency calibration. Each channel of this microscope achieves a spectral resolution of ΔE/E ∼ 10%, and the efficiency uniformity (relative standard deviation: ση/η̄) within the ±150 μm field of view is 6.27%–9.63%. The spatial resolution at the center of the field of view is better than 5 μm. We deployed this microscope for implosion hotspot imaging, successfully acquiring images at four distinct energy points. These images are suitable for diagnosing hotspot electron temperature.
X-ray security inspection systems face significant challenges when deployed across different scanning devices due to domain shifts that substantially degrade prohibited item recognition performance. Existing deep learning approaches assume consistent label distributions between training and testing domains, which is impractical in real-world security scenarios where devices vary in penetration capabilities and imaging characteristics. To address this critical limitation, we propose a Weighted Adaptive Feature Adversarial Network (WAFAN) that leverages partial domain adaptation (PDA) to achieve robust cross-domain prohibited item recognition. Our approach introduces three key innovations: (1) a similarity-based weight assignment strategy that effectively distinguishes relevant target-domain classes from irrelevant source-domain categories, mitigating negative transfer effects; (2) an adaptive convolution module integrated via residual connections that explicitly captures and compensates for cross-domain feature distribution differences; and (3) a multi-space adversarial learning framework that ensures domain-invariant feature extraction. Comprehensive experiments on PIDray and SIXray datasets demonstrate that WAFAN achieves superior performance over state-of-the-art PDA methods, with classification accuracy exceeding 80% on target domains while maintaining computational efficiency. The proposed method offers a practical solution for deploying AI-powered security inspection systems across heterogeneous X-ray devices without requiring extensive retraining or domain-specific annotations.
High-energy X-ray diagnostic systems are crucial for understanding hotspot high-density area asymmetry, fuel mixing, and other phenomena in inertial confinement fusion. To meet the demand for hotspot electron temperature measurements, we developed a high-energy dual-channel Kirkpatrick–Baez microscope. This microscope is characterized by a dual high-energy response and high spatial resolution, enabling the observation of fine structures in high-density regions of a hotspot. Spectral drift was effectively mitigated by optimizing the grazing incidence angle, and the spatial and spectral domains were coupled through experimental alignment. Herein, we describe the optical design of the proposed microscope. Furthermore, we performed simulations and backlight imaging to validate the performance of the proposed system. The results show that the spatial resolution was better than 3 μm in the center and better than 6.5 μm in a field of view of 300 μm. The spectral response efficiencies at 11.4 and 17.48 keV were 7.41 × 10−8 and 5.77 × 10−8 sr, which deviate from the theoretical values by 3.01% and 6.79%, respectively.
9 March 2026 in IEEE Transactions on Information Forensics and Security
With increasing security threats in critical infrastructure, intelligent X-ray inspection systems have become essential for modern security frameworks. Contraband detection faces dual challenges: semantic dilution across cross-scale features and degraded directional sensitivity. Existing detection paradigms rely on Euclidean spatial assumptions while ignoring X-ray projection geometry characteristics, leading to feature representation ambiguity and insufficient spatial relationship modeling in complex occlusion scenarios. To address these challenges, we propose RFFRDet, a rotation detector based on refined feature decoupling. First, an Integrated Local Attention (ILA) module is constructed that enables material-aware and geometry-aware feature enhancement through channel-spatial decoupling. Second, a Multi-Resolution Feature Fusion (MRFF) network is designed that achieves optimal coupling between fine-grained spatial positioning and high-level semantic understanding through parallel multi-scale aggregation. Finally, we construct the first multi-directional X-ray security benchmark dataset with rotation annotations for 27,000 images. Extensive experiments demonstrate that RFFRDet achieves 97.4% mAR and 96.8% mAP, representing improvements of 1.6% and 1.5% over state-of-the-art methods, respectively.
1 November 2026 in Review of Scientific Instruments
Quantitative analysis of inertial confinement fusion (ICF) implosion hotspots requires accurate measurements of spatially distributed x-ray emission spectra. Kirkpatrick–Baez (KB) x-ray microscopes are widely used for hotspot diagnostics in major ICF facilities; however, accurate calibration of their full-field and full-spectrum response efficiency remains challenging due to multi-parameter coupling and spatially nonuniform optical response. In this study, we present an efficient calibration method for a KB microscope by combining theoretical modeling with experimental reflectivity measurements. The response efficiency is described in terms of the collection solid angle, magnification, and angle-dependent mirror reflectivity, while experimental data are used to calibrate key model parameters, including the effective grazing incidence angle and a global intensity correction factor accounting for system-level losses. The method is applied to a four-channel, energy-adaptive KB microscope, enabling determination of the spatial distribution of the effective response energy and the full-spectrum response efficiency over the entire field of view. This approach provides a reliable basis for quantitative ICF hotspot imaging diagnostics.