倾斜影像区块化自动整理方法研究与实现

高精度倾斜摄影实景三维重建,基于区块化数据规范管理、像控制点选片、倾斜影像姿态信息整理等需求,在数据导入建模软件之前,需要对海量倾斜影像数据进行人工整理。传统倾斜摄影后续数据整理,涉及影像分类、按规范重命名、影像按航向旋转、倾斜影像信息表录入等繁杂的工序,通常需要人工进行操作,劳动强度大,特别是针对多视角海量数据,错误率较高。该文旨在研究一种基于消费级无人机的高精度倾斜摄影区块化倾斜影像快速自动整理方法,通过程序替代人工批量实现倾斜影像视角分类、多视角影像重命名、Exif-GPS信息提取、曝光点轨迹及区块化倾斜影像信息表制作等工作,并以山东省滕州市前杨岗村高精度实景三维重建为实例,验证该方法能够极大降低人工劳动强度、缩短数据整理周期、保障数据精度,对提高基于消费级无人机倾斜摄影的实景三维建模效率具有较大的借鉴及实用价值。

Research and Implementation of Automatic Sorting Method for Tilted Image Block

Based on the requirement of block data management, control points selection and tilt image attitude information collation, a large amount of tilt image data need to be manually collated before data is imported into the modeling software. The follow-up data processing of traditional tilt photography involves complicated processes, such as image classification, renaming according to specifications, image rotation according to heading, input of tilt image information table, etc. which usually requires manual operation with high labor intensity, especially for multi view massive data with high error rate. The purpose of this paper is to study a fast automatic sorting method of tilt images based on consumer UAV. The program replaces manual batch to realize classification of tilt images, multi-view images renaming, Exif-GPS information extraction, exposure points trajectory and block tilt image information table making. The real scene of Qianyanggang village in Tengzhou city of Shandong province is taken as an example to prove that this method can greatly reduce the labor intensity, shorten the data processing cycle. It is of great reference and practical value to improve the efficiency of real scene 3D modeling based on consumer UAV tilt photography.