Increasing the transmittance of zinc sulfide (ZnS) infrared windows can effectively improve the imaging quality of infrared detection. In this study, an anti-reflective subwavelength structure (ASS) was manufactured on ZnS using a femtosecond burst Bessel laser with the goal of achieving high transmittance in the mid-infrared range. The period and depth parameters of the ASS were initially determined using the effective medium approximation (EMA) theory and subsequently optimized using the rigorous coupled-wave analysis (RCWA) method to eliminate surface Fresnel anti-reflections. The depth of the ASS increases with the number of bursts, while the structure profile transitions from Gaussian to conical. In addition, the ASS achieves 86% transmittance in the 7–10 µm range, and the average transmittance improves by 10% in the 5–12 µm range. Moreover, the wide-angle ASS with the hydrophobicity (contact angle 160°) is achieved on the ZnS window. Ultimately, the ASS on ZnS enhances the clarity of the infrared image.

Accurate and efficient detection of small and complex surface defects on hot-rolled steel plates remains a significant challenge in industrial quality assurance. Current deep learning detectors often exhibit limitations in detection accuracy and training convergence speed, particularly for small objects, which limits their practical deployment in real-time industrial inspection systems. To overcome these deficiencies, this paper proposes a multi-scale deformable transformer iterative query refinement network (MDT-Net). MDT integrates three key innovations: a Swin Transformer backbone for robust multi-scale feature representation, a deformable attention mechanism to significantly reduce computational complexity and accelerate convergence, and an iterative bounding box refinement strategy for precise localization. Extensive experiments on the NEU-DET dataset demonstrate MDT’s superior performance, achieving 82.7% mAP₅₀. Crucially, MDT significantly outperforms other mainstream detectors in small object detection, reaching an mAP_50:95 of 0.55, and exhibits remarkably faster training convergence. These findings confirm MDT’s effectiveness and robustness for accurate and efficient steel surface defect detection, thereby providing a crucial tool for enhancing sensor-based quality control and offering a promising solution for industrial quality management.