Accurate weld keypoint localization is an important visual perception task for robotic welding, weld tracking, and intelligent manufacturing. However, weld keypoints in line-structured light images are usually small, weakly textured, and easily affected by reflections, noise, and spurious laser stripes. These factors make it difficult for lightweight detection models to balance localization accuracy and computational efficiency. To address this problem, this paper proposes Weld-FHG-YOLO, a lightweight multi-frequency feature fusion network for weld keypoint localization. The proposed model is built on the You Only Look Once version 11 nano (YOLO11n) framework and focuses on optimizing feature fusion and scale transformation in the Neck. Specifically, FasterC3K2 is introduced to replace the original C3K2 modules in the Neck, thereby reducing redundant computation during multi-scale feature fusion. In addition, a Haar Wavelet Decomposition and Group Shuffle Convolution (HWD-GSConv) downsampling fusion module is designed, in which Haar wavelet decomposition preserves low-frequency structural information and high-frequency details, while GSConv performs lightweight multi-frequency feature fusion. Experimental results show that Weld-FHG-YOLO achieves 2.301 M parameters and 6.016 GFLOPs, which are 11.26% and 6.83% lower than those of YOLO11n, respectively. Meanwhile, mAP@0.5:0.95 increases from 0.7337 to 0.7901, the Mean Center Error (MCE) decreases from 2.254 px to 2.131 px, and the CPU inference speed increases from 13.69 to 14.77 frames per second (FPS). These results indicate that the proposed method improves strict localization accuracy and localization stability while maintaining a lightweight computational profile, providing a practical visual perception approach for weld keypoint localization in resource-constrained intelligent manufacturing scenarios.

Rice pest detection in natural field images is affected by weak visual cues and interference from complex crop backgrounds. To improve the use of local appearance and surrounding contextual information, this study proposes LCIC-YOLO, a lightweight detector based on YOLO11n. An ICGSimConv module is introduced into the high-resolution $P_3$ pathway to extract Local and Context features and regulate them independently using energy-based spatial responses before fusion. A six-class RP6 dataset containing 2637 images and 4312 annotated instances was curated from multiple public sources using duplicate detection and visual-family-based partitioning. Across three training seeds under the RP6 evaluation protocol, LCIC-YOLO achieved an mAP${}_{50}$ of 0.8826 ± 0.0051 and an mAP${}_{50--95}$ of 0.5902 ± 0.0047, with 2.60 M parameters and 6.57 GFLOPs. Its mean mAP${}_{50--95}$ exceeded YOLO11n by 2.67 percentage points. The control with matched gating and the fusion ablations support the contribution of Local information when integrated through branch-specific response estimation and independent Local–Context regulation. These results indicate that LCIC-YOLO improves the use of complementary spatial information while retaining the lightweight characteristics of the baseline detector.