Abstract
Abstract. Automatic image registration is a basic step in multi-sensor data integration in remote sensing and photogrammetric applications such as data fusion. The effectiveness of Mutual Information (MI) as a technique for automated multi-sensor image registration has previously been demonstrated for medical and remote sensing applications. In this paper, a new General Weighted MI (GWMI) approach that improves the robustness of MI to local maxima, particularly in the case of registering optical imagery and 3D point clouds, is presented. Two different methods including a Gaussian Mixture Model (GMM) and Kernel Density Estimation have been used to define the weight function of joint probability, regardless of the modality of the data being registered. The Expectation Maximizing method is then used to estimate parameters of GMM, and in order to reduce the cost of computation, a multi-resolution strategy has been used. The performance of the proposed GWMI method for the registration of aerial orthotoimagery and LiDAR range and intensity information has been experimentally evaluated and the results obtained are presented.
Highlights
The automatic registration of multi-sensor remote sensing data has generated much research interest in remote sensing and digital photogrammetry
The registration of the orthoimagery and LiDAR data sets is impacted by the following factors: Firstly, they were acquired a year apart and changes had occurred within the scene, mainly in isolated buildings, and to an extent in vegetation and roads
The ability of the developed weighted mutual information-based registration approach to efficiently register multi-sensor data has been highlighted in this paper
Summary
The automatic registration of multi-sensor remote sensing data has generated much research interest in remote sensing and digital photogrammetry. Chung et al (2007) present a method based on a priori knowledge of the joint intensity distribution between image pairs at different image resolutions and the Kullback-Leibler (KLD) distance similarity measure to achieve superior robustness in multimodal image registration. Contrary to method proposed in this paper, those referred to above do not define a general weighting function for the joint intensity distribution in registration of data from multiple sensors. They do not provide an approach which is independent of prior knowledge in the estimation of the joint intensity distribution to remove the effect of local maxima in the registration result. In order to efficiently accommodate the registration of multi-sensor data, including airborne and space-borne imagery and ranging data, a novel method called General Weighted Mutual Information (GWMI) has been developed
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
