The invention discloses a method and a device for identifying inclined pulling and inclined hoisting of a crane. The method comprises the following steps: collecting picture data of a lifting hook through a camera, and training a target detection model; the method comprises the following steps: recording standard coordinates of a lifting hook in a free falling state through data calibration when different boom inclination angles x and lifting hook heights y are recorded, and establishing a curved surface model; acquiring a real-time image data stream of the lifting hook; calling the target detection model to obtain real coordinates of the lifting hook; acquiring a real-time current boom inclination angle x and a lifting hook height y of the lifting hook; calling the curved surface model to obtain standard coordinates of the lifting hook in a free falling body state; calculating the distance between the real coordinate and the standard coordinate, and if the distance is greater than a threshold value, determining that the crane is obliquely pulled and lifted; the method has no requirement on the position of the camera mounted on the arm support, can be mounted at any position near an olecranon of the arm support, is high in speed and can be used for real-time identification. The precision is high, and the coordinate position of the lifting hook and the coordinate position of a standard point can be accurately obtained in a complex scene.
本发明公开一种用于起重机识别歪拉斜吊的方法及装置,本发明通过摄像头采集吊钩图片数据,训练目标检测模型;通过数据标定,记录不同臂架倾角x以及吊钩高度y时,获得的吊钩自由落体状态下的标准坐标,建立曲面模型;获取吊钩的实时图像数据流;调用目标检测模型获取吊钩的真实坐标;获取吊钩的实时当前臂架倾角x,和吊钩高度y;调用曲面模型得到自由落体状态下吊钩的标准坐标;计算真实坐标与标准坐标之间的距离,如果大于阈值,确定起重机存在歪拉斜吊;本发明对摄像头安装在臂架的位置没有要求,可以安装在臂架鹰嘴附近任何位置,速度快,可以实时进行识别。精度高,在复杂场景下可以准确获得吊钩坐标位置,已经标准点坐标位置。
Method and device for identifying inclined pulling and inclined hoisting of crane
一种用于起重机识别歪拉斜吊的方法及装置
2024-03-01
Patent
Elektronische Ressource
Chinesisch
Self-correcting hoisting method for inclined pulling and inclined hoisting of translation mechanism
Europäisches Patentamt | 2023
|