全景深度图像精度分析及优化方法研究

针对三维激光点云数据生成360°全景深度图像存在像素分辨率不均匀的问题,提出一种顾及目标量测精度及可见度的全景深度图像生成方法,在保证全景影像表达地物的完整性的前提下提高其数据的存取精度。通过坐标转换和投影变换生成与全景影像匹配的全景深度图像;基于摄影成像原理分析摄影中心高度、深度值和像素分辨率之间的关系,得到不同深度处目标分辨率随摄影中心升高趋于一致的结论;综合分析地面目标分辨率和杆状目标尤其是树冠对树干的遮挡问题,确定特定场景下生成全景深度图像的最佳摄影中心位置并重新生成深度图像。实验分析表明,该方法能够在保证杆目标可见度的前提下提高地面目标量测精度。

Accuracy analysis and optimization of panoramic depth image

The pixel resolution of 360°panoramic depth image generated by 3Dlaser point cloud data is uneven.A panoramic depth image generation method considering the measurement accuracy and visibility of the target is proposed to improve the access accuracy of the data under the premise of ensuring the integrity of the panoramic image expressing the ground objects.Firstly,panoramic depth image matching panoramic image is generated by coordinate transformation and projection transformation.Then,based on the principle of photographic imaging,the relationship between the height,depth and pixel resolution of the photographic center is analyzed,and the conclusion that the target resolution at different depths tends to be consistent with the increase of the photographic center is obtained.Finally,the resolution of the ground target and the occlusion of the rod target,especially the crown,to the tree trunk are comprehensively analyzed,and the optimal location of the photographic center for generating panoramic depth image in a specific scene is determined and the depth image is regenerated.The experimental analysis shows that the method can improve the measurement accuracy of ground target under the premise of ensuring the visibility of rod target.