【摘要】 针对易受环境影响、无法全面覆盖作业区域等传统速度检测方法的局限性，提出一种液压平板车的速度检测方法。该方法利用无人机航拍的视频制作液压平板车的数据集，在对待检测视频进行降噪处理并且确定测速区域后，使用深度学习算法(YOLO)识别并跟踪液压平板车，对处理后的视频开展仿真实验，实现对液压平板车速度的检测。实验结果表明，该方法测得的速度与实际速度相比误差较低，为船厂检测液压平板车行驶速度提供切实可行的方法。更多还原

【Abstract】 Aiming at the limitations of traditional speed detection methods, such as being susceptible to environmental influences and not being able to fully cover the operation area, a speed detection method for hydraulic flatbed trucks was proposed. The method made use of UAV aerial video to create a dataset of hydraulic flatbed trucks, and after noise reduction processing of the video to be detected and determining the speed measurement area, the deep learning algorithm YOLO(you only look once) was used to recognize and track the hydraulic flatbed trucks, and finally simulation experiments were carried out on the processed video. The experimental results showed that the speed measured by this method has a low error compared with the actual speed, which provides a practical method for shipyards to detect the traveling speed of hydraulic flatbed trucks.更多还原