The inspection of welding surface quality is an important task for welding work. With the development of product quality inspection technology, automated and machine vision-based inspection have been applied to more industrial application fields because of its non-contact, convenience, and high efficiency. However, challenging material and optical phenomena such as high reflective surface areas often present on welding seams tend to produce artifacts such as holes in the reconstructed model using current visual sensors, hence leading to insufficiency or even errors in the inspection result. This paper presents a 3D reconstruction technique for highly reflective welding surfaces based on binocular style structured light stereo vision. The method starts from capturing a fully lit image for identifying highly reflective regions on a welding surface using conventional computer vision models, including gray-scale, binarization, dilation, and erosion. Then, fringe projection profilometry is used to generate point clouds on the interested area. The mapping and alignment from 2D image to 3D point cloud is then established to highlight features that are vital for eliminating “holes”—large featureless areas—caused by high reflections such as the specular mirroring effect. A two-way slicing method is proposed to operate on the refined point cloud, following the concept of dimensionality reduction to project the sliced point cloud onto different image planes before a Smoothing Spline model is applied to fit the discrete point formed by projection. The 3D coordinate values of points in the “hole” region are estimated according to the fitted curves and appended to the original point cloud using iterative algorithms. Experiment results verify that the proposed method can accurately reconstruct a wide range of welding surfaces with significantly improved precision.

Bubble defects are a common type of tire quality defect, which are difficult to detect directly with the naked eye or using X-ray methods. This paper employs electronic shearography to non-destructively detect internal bubble defects in tires. Addressing the issues of low efficiency and poor accuracy in manual image recognition, as well as the inability to assess bubble defect sizes, a machine vision-based method for detecting tire bubble defects is proposed. Utilizing sine-cosine transform, least squares phase unwrapping, contrast-based and RANSAC edge detection algorithms, combined with deep learning techniques, this method achieves a comprehensive solution from automatic identification to false detection screening, and then to the assessment of bubble defect sizes based on out-of-plane displacement, providing a quality rating for tires. Experimental validation on actual samples demonstrates that the method achieves a missed detection rate and false detection rate of less than 2%, meeting the required accuracy standards for detection.

Reliable detection of X-ray tire defects is essential for safety and quality assurance in manufacturing. However, low contrast and high noise make traditional methods unreliable. This paper presents DyReCS-YOLO, a dynamic re-parameterized channel-shuffle network based on YOLOv8. The model introduces a C2f_DyRepFusion module combining dynamic convolution and a shuffle-and-routing mechanism, enabling adaptive kernel adjustment and efficient cross-channel interaction. Experiments on an industrial X-ray tire dataset containing 8326 images across 58 defect categories demonstrate that DyReCS-YOLO achieves an mAP@0.5 of 0.741 and mAP@0.5:0.95 of 0.505, representing improvements of 4.5 and 2.8 percentage points over YOLOv8-s, and 9.2 and 7.7 percentage points over YOLOv11-s, respectively. The precision increases from 0.698 (YOLOv8-s) and 0.668 (YOLOv11-s) to 0.739, while maintaining real-time inference at 189.5 FPS, meeting industrial online detection requirements. Ablation results confirm that the combination of dynamic convolution and channel shuffle improves small-defect perception and robustness. Moreover, DyReCS-YOLO achieves an mAP@0.5 of 0.975 on the public MT defect dataset, verifying its strong cross-domain generalization.

For the harsh environment and uneven ground during excavator slope repair operations, and to solve the problem of restricted vision of the driver during manual operation, a slope information perception method based on binocular cameras and auxiliary lines along the lower edge of the slope is proposed. The slope and the surrounding environment are matched in three dimensions by the binocular camera and the point cloud is reconstructed, and the auxiliary line point cloud is extracted by point cloud segmentation according to the difference in colour of the point cloud, and fitted to a spatial straight line as the position where the bucket stops working. For the typical working conditions of excavator slope repair construction, the most influential factors were selected for comparison experiments, and it was verified that the perception error of slope-related information by this method is within 4cm. The method can provide accurate information on the distance and coordinates of the slope and the auxiliary line, which improves the disadvantages of manual operation and provides information support for the autonomous slope repair of excavators.