Weld feature point detection is a key technology for welding trajectory planning and tracking. Existing two-stage detection methods and conventional convolutional neural network (CNN)-based approaches encounter performance bottlenecks under extreme welding noise conditions. To better obtain accurate weld feature point locations in high-noise environments, we propose a feature point detection network, YOLO-Weld, based on an improved You Only Look Once version 5 (YOLOv5). By introducing the reparameterized convolutional neural network (RepVGG) module, the network structure is optimized, enhancing detection speed. The utilization of a normalization-based attention module (NAM) in the network enhances the network’s perception of feature points. A lightweight decoupled head, RD-Head, is designed to improve classification and regression accuracy. Furthermore, a welding noise generation method is proposed, increasing the model’s robustness in extreme noise environments. Finally, the model is tested on a custom dataset of five weld types, demonstrating better performance than two-stage detection methods and conventional CNN approaches. The proposed model can accurately detect feature points in high-noise environments while meeting real-time welding requirements. In terms of the model’s performance, the average error of detecting feature points in images is 2.100 pixels, while the average error in the world coordinate system is 0.114 mm, sufficiently meeting the accuracy needs of various practical welding tasks.

Hybrid refractive–metalens zoom systems offer a promising route to overcome the size, mass, power consumption, cost, and response-time limitations of conventional mechanically compensated refractive zoom systems by exploiting the high degrees of freedom in wavefront manipulation provided by tunable metalenses. A Seidel-compatible initial-structure calculation method is presented for axially fixed hybrid zoom optical systems. The method applies to architectures containing two tunable metalens groups and fixed refractive elements arranged arbitrarily. By treating the Seidel aberrations of the tunable metalens groups as known quantities in an extended Mikš third-order aberration matrix, we reformulate the inverse determination of the fixed refractive-element parameters as a linear least-squares problem. This approach yields physically interpretable and aberration-compatible initial structures, which can be directly imported into optical design software such as Zemax. Case studies on two classical configurations—Alvarez and Moiré—demonstrate that the method enables starting configuration calculation for hybrid refractive–metalens zoom optical systems, transforming the configuration search from empirical trial-and-error into a physics-driven parameter-determination problem and thereby providing a physically interpretable starting point for subsequent image quality optimization.