Search

This thesis details the development of a mobile mapping system (MMS) using smartphones and investigates its ability to be a major source for Geographic Information System (GIS) and mapping applications. Nowadays, smartphones contain all the required components for any MMS (i.e., Global Positioning System (GPS) receiver, a digital camera, accelerometers, gyroscopes, and magnetometers). However, the accuracy of the sensors in smartphones is poor, which is the main drawback of using smartphones for mapping applications. In this thesis, two Android applications were developed to capture synchronized images or videos with various sensors measurements. Using a smartphone’s GPS receiver, its initial positions at the instants of exposure can be estimated; and a smartphone’s initial orientation parameters can be calculated using its accelerometers and magnetometers sensors. These initial values, which are called the Exterior Orientation Parameters (EOPs) and the initial Interior Orientation Parameters (IOPs) of the camera in use, are corrected using epipolar geometry constraints along with a set of automatically matched points between each two consecutive images. The corrected EOPs and IOPs are then used to estimate the 3D coordinates of the interest points using a developed bundle adjustment software application. The implementation of the prototype system is also shown, and all of the methods utilized to obtain the final mapping results are detailed. Using this prototype system, examples of the mapping results are shown where the absolute maximum mapping errors did not exceed 0.32 and 0.5 meter in the horizontal and vertical directions respectively using six captured images and two control points. Other obtained mapping results also are presented in this thesis to demonstrate the developed system’s ability to obtain mapping results efficiently, by pedestrians or from a moving vehicle, using both the captured images and recorded videos.