The Research On Multi-Sensors Calibration Of Mobile Mapping Systems

Among the information obtained from outside environment, most of them come from human visual perception. The information is delivered from eyes to brain, then analyzed and processed by brain. With the development of the sensors and computer technology, computer vision, which subsitute humans’cerebra and eyes, can achieve the simple function of environmental recognition and understanding. The aim of Computer vision research is to make the computer enable to "understand" a scene like a man. This capability refers to not only the objects’geometrical information perception which includes the information of shape, position, pose and motion etc., but also information describing, storing, recognizing and understanding. Compared with obtaining the3D surrounding information through different kinds of machinery measurement tools, angle measurement and distance measurement tools, the method based on digital photogrammetry could obtain the targets’instant information in a particular motion state, so it has the advantages inculding real time, effecient, accuracy, conceal etc.Mobile mapping system is the dynamic integration of digital photogrammetry and located multiple sensors. A mobile mapping system consists mainly of a moving platform, navigation sensors, and mapping sensors. Generally, the mobile mapping system choose a land vehicle, a vessel, or an aircraft as the moving platform, combining the mapping sensors (such as charge coupled devices (CCD) cameras, laser rangers etc.,)and navigation sensors (for instance Global Positioning System (GPS), vehicle wheel sensors, and inertial navigation systems (INS) etc.,)to obtain high flexibility in data acquisition, more information with less time and effort. Which significant avoid the disadvantages of the existing measuring methods.This paper proposes a calibration program base on multiple sensors of mobile mapping system, which includes cameras, INS and GPS.Based on the research of classical camera calibration methods, the paper propose a novel camera calibration method which could calibrate the internal and external parameters by single image of mulple control points. And to further realize the solution of corresponding points coordinates by forward intersection with two cameras.Besides, the paper proposes a design program of calibration field and a feature point extraction method. Moreover, it proposes a design program of the3D target and researches the algorithms of feature point extraction in3D point clouds.It proposes an algorithm of feature point extraction in3D point clouds by the modified generalized projection based M-estimator. Through the study of this paper, it solves the crucial technology of multi-sensors calibration of mobile mapping system and makes the whole technological process smooth realization.