基于深度传感器的煤场三维测量改进算法研究

Research on Improved Algorithm of Three-dimensional Measurement in Coal Yard Based on Depth Sensor

针对传统盘煤仪使用过程中存在的标定误差大、模型重建失效等问题,提出基于直接线性变换模型的改进标定算法、基于飞行时间的改进三维重建算法,通过固定式深度传感器与相机实现盘煤。同时,针对盘煤仪无法可视化、体积测量误差大等问题,提出基于Unity3D的虚拟场景设计、基于颜色和法向量的点云分割方法,实现煤场三维可视化,从而计算模型体积。经实验验证,改进算法测量精度高、扫描速度快、设备稳定性强,具有较高使用价值。

In order to solve the problems of large calibration error and model reconstruction failure in the operation of traditional coaling instrument,an improved calibration algorithm based on direct linear transformation model and an improved 3D reconstruction algorithm based on time-of-flight method were proposed.The coaling was realized by a fixed depth sensor and a camera.At the same time,in view of the problems such as the lack of visualization of the coaling instrument and large volume measurement error,the virtual scene design based on Unity3D and the point cloud segmentation method based on color and normal vector were proposed to realize the three-dimensional visualization of the coal field and calculate the model volume.The experimental results show that the improved algorithm has high measuring accuracy,fast scanning speed and strong stability of the equipment,so it has high application value.