Datasets of captured images of three different devices for photogrammetry calculation comparison and integration into a laserscan point cloud of a built environment

The presented data article aims to provide the whole dataset obtained during an experiment of updating laser scan point clouds with photogrammetry meshes. In this context, the data quality and calculation time of photogrammetry models from different recording devices and different software solutions were compared. It was investigated whether photos from smartphones are also appropriate for updating point clouds by using photogrammetry in a factory environment. The photos of a technical installation were taken in 08:30 min with these three devices: Nikon D810 with Sigma art 24mm, iPhone 6 and iPhone XS. With each of the mentioned devices, three datasets have been created to provide enough data for the comparisons. One dataset (photos in .TIFF) of the iPhone XS is provided. The results of the data sets are used for a photogrammetry mesh quality comparison and a calculation time comparison. For the mesh quality comparison, visual qualitative inspections were performed on the models and the results were compared. Furthermore, all settings in the RealityCapture BETA 1.0.3.9696 ppi and Meshroom 2019 2.0 software are provided. A comparison of the quality of the photogrammetric 3D meshes was performed by comparing the rendering results. The dataset of the iPhone XS can be used to compare further photogrammetry software or single algorithms. Besides the images, the initial point cloud of the laser scanner is provided. Also included is the combined file which consists of the laser scan point cloud and the photogrammetry mesh of the end of the experiment.

本数据论文旨在提供一项实验的完整数据集，该实验通过摄影测量网格（photogrammetry meshes）更新激光扫描点云（laser scan point clouds）。在此研究框架下，本研究对比了不同记录设备与不同软件方案生成的摄影测量模型的数据质量与计算耗时。本研究旨在探究在工厂环境中，使用摄影测量技术将智能手机拍摄的照片用于更新点云是否具备可行性。技术装置的照片于08:30时段使用以下三种设备拍摄：搭载Sigma Art 24mm镜头的尼康D810、iPhone 6以及iPhone XS。每种设备均生成三组数据集，以提供充足的对比样本数据。本次公开了iPhone XS拍摄的.TIFF格式照片数据集。各数据集的分析结果可用于摄影测量网格质量对比与计算耗时对比研究。针对网格质量对比环节，我们对生成的模型开展了视觉定性评估，并对各模型的评估结果进行了横向比对。此外，本次公开了RealityCapture BETA 1.0.3.9696 ppi与Meshroom 2019 2.0两款摄影测量软件中的全部设置参数。通过比对各模型的渲染结果，完成了摄影测量三维网格的质量对比分析。iPhone XS的数据集可用于后续针对其他摄影测量软件或单一算法的对比测试。除图像数据外，本次公开了激光扫描仪采集得到的初始点云数据集。同时公开了实验结束时，由激光扫描点云与摄影测量网格融合得到的合成文件。