Abstract: Aiming at the problems of non-uniform motion blur, illumination changes, and background interference in the detection for load swing angle of bridge cranes, a real-time monocular vision-based method for load swing angle detection is proposed. This method utilizes spherical markers as feature targets and constructs a collaborative framework integrating a Channel and Spatial Reliability Tracker with Adaptive Feature Tracking (CSRT-AFT) and a gradient-layered adaptive circle detection algorithm. The CSRT-AFT achieves robust tracking through a dynamic trajectory filtering and feature adaptation mechanism. This includes an adaptive multi-modal trajectory filtering strategy, intelligently switched based on curvature change rate and acceleration mutation index to suppress severe motion trajectory jitter. It also combines dynamic Oriented FAST and Rotated BRIEF (ORB) feature extraction, weighted K-Dimensional (K-D) feature screening, and the least median of squares method for feature matching and elastic template updating, enhancing feature matching stability under motion blur and complex illumination. For rapid and precise target localization, building upon image preprocessing for enhanced robustness, the gradient-layered adaptive circle detection algorithm achieves sub-pixel level fast circle detection through gradient field-guided circle center candidate generation, multi-stage probability sampling, and geometric constraint verification. Finally, a load swing angle measurement model is established in conjunction with the bridge crane's workspace. Experimental results demonstrate that this method can stably detect the load target's swing angle under various trolley speeds and complex conditions including illumination changes and occlusions, significantly improving detection accuracy and real-time performance.

Key words: bridge crane, monocular vision, adaptive algorithm, image processing, swing angle measurement, blurred images

摘要： 针对桥式起重机负载摆角检测中非匀速运动模糊、光照变化及背景干扰问题，提出一种基于单目视觉负载摆角实时检测方法。该方法以球形标记为特征目标，构建通道与空间可靠性自适应特征跟踪器（CSRT-AFT）与梯度分层自适应圆检测算法协同框架。CSRT-AFT通过动态轨迹滤波与特征自适应机制实现鲁棒跟踪：设计自适应多模态轨迹滤波，基于曲率变化率与加速度突变指数智能切换滤波策略，抑制剧烈运动轨迹抖动；结合动态定向快速和旋转BRIEF（ORB）特征提取、加权K维（K-D）特征筛选、最小中值平均法进行特征匹配及弹性模板更新，增强运动模糊和复杂光照下特征匹配稳定性。为实现目标的快速精确定位，在图像预处理增强鲁棒性的基础上，梯度分层自适应圆检测算法通过梯度场引导圆心候选生成、多阶段概率采样及几何约束验证，实现亚像素级的快速圆检测。最后结合桥式起重机工作空间建立负载摆角测量模型。实验结果表明，该方法在不同小车运动速度和复杂光照、遮挡条件下均能稳定检测负载目标摆角，大幅提升了检测的精度与实时性。

关键词: 桥式起重机, 单目视觉, 自适应算法, 图像处理, 摆角测量, 模糊图像