Integrierte Georeferenzierung für bildgebende Systeme mit sequentieller Datenakquisition

具有顺序数据采集功能的成像系统集成地理配准

基本信息

批准号：
223861662
负责人：
Professor Dr. Hans-Gerd Maas
金额：
--
依托单位：
Institut für Photogrammetrie und Fernerkundung
依托单位国家：
德国
项目类别：
Research Grants
财政年份：
2012
资助国家：
德国
起止时间：
2011-12-31 至 2016-12-31
项目状态：
已结题

来源：
https://gepris.dfg.de/gepris/projekt/223861662?language=en
关键词：
Integrierte Georeferenzierung bildgebende Systeme mit

项目摘要

The goal of the project is the development of methods for an integrated georeferencing of laserscanner and camera data acquired from a moving platform. While the processing of the laserscanner and camera data allows for internal centimeter-accuracy, the vehicle position and attitude sensorics will often be much less precise and will thus not allow for adequate georeferencing. This is typical for areas such as mobile robotics or UAVs (unmanned aerial vehicles). The georeferencing process is further impeded by high requirements concerning the temporal resolution of the trajectory representation due to the sequential manner of laserscanner data acquisition.Therefore, the platform position and orientation parameters shall be determined by an integrated bundle adjustment procedure. Due to the lack of temporal resolution, the bundle adjustment cannot purely be based on the camera images. Instead, a time-resolved parameterized platform trajectory is introduced as an unknown into the adjustment. The trajectory determination is supported by introducing laserscanner measurements in the integrated bundle adjustment. Integrating image and laserscanner data in a bundle adjustment procedure also requires the development of techniques for the determination of homologous features in the image and laserscanner data as well as the development of stochastic models for an adaptive balancing of the weights of the different observation classes.

该项目的目的是开发从移动平台中获得的Laserscanner和相机数据集成的地理发射的方法。尽管LaserScanner和相机数据的处理允许内部厘米精度，但车辆位置和态度传感器通常会少得多，因此不允许进行足够的地理处理。这对于移动机器人技术或无人机（无人驾驶飞机）等领域是典型的。由于LaserScanner数据采集的顺序方式，关于轨迹表示的时间分辨率的高要求进一步阻碍了地理发行过程。因此，平台位置和方向参数应由集成的捆绑包调整程序确定。由于缺乏时间分辨率，因此捆绑调整不能完全基于相机图像。取而代之的是，将时间分辨的参数化平台轨迹引入了调整中。通过在集成束调整中引入LaserScanner测量值来支持轨迹的确定。在捆绑调整过程中将图像和Laserscanner数据集成在一起，还需要开发技术，以确定图像和Laserscanner数据中的同源特征，以及开发随机模型，以自适应平衡不同观察类别的重量。