The orthophoto with refined details and higher accuracy is important for urban geographical information system. The traditional differential rectification did not consider the height information of buildings when dealing with imagery over urban areas, resulting in buildings having relief displacement and cannot be located at their true geographical positions. In this study, a digital building model (DBM)-based procedure for automatic true orthophoto generation is proposed to solve this problem. This procedure includes three major steps: (1) traditional orthophotos generation, (2) buildings relief correction, and (3) occlusion detection and compensation. In our method, the relief displacements for buildings are corrected and occlusions are detected by using the backprojection and intersection method based on vector DBM surface polygon. True orthophotos are obtained with the compensation of occlusions. Experimental results show that the generated true orthophotos can achieve root-mean-square errors of 0.149 and 0.061 m on the X- and Y-axes, respectively. The planimetric positioning accuracy of the true orthophoto is around 1 pixel. This indicates that the proposed method can correctly remove the displacement caused by terrain and tall buildings, and the occluded areas can be detected and compensated effectively for generating true orthophotos with high quality.
This paper describes an algorithm framework for registration of airborne based laser scanning data (LIDAR) and optical images by using multiple types of geometric features. The technique utilizes 2D/3D correspondences between points and lines, and it could easily be extended to general features. In generalized point photogrammetry, all lines and curves are consists of points, which could be describe in collinear equation, so it could represent all kinds of homogeneous features in an uniform framework. For many overlapping images in a block, the images are registered to the laser data by the hybrid block adjustment based on an integrated optimization procedure. In addition to the theoretical method , the paper presents a experimental analysis the sensitivity and robustness of this approach
This paper describes an algorithm framework for fusion airborne based laser scanning data (LIDAR) and optical images.
A efficient and reliable intensity-based registration framework has been used to determining the spatial transform form
LIDAR to optical images. On the basis of segmented airborne images, the paper raises the arithmetic and process of
merging multi-data sources to carry through classification by using multi-echo, point's space discrete characteristics, and
the statistic spectrum characteristics. In addition to the theoretical method, the paper presents a experimental analysis the
sensitivity and robustness of this approach to assess effectiveness the proposed arithmetic.
This paper describes an algorithm framework for automatic registration of airborne based laser scanning data (LIDAR)
and optical images by using mutual information. The part on methodology describes aspects such as pre-processing of
images, intensity value interpolation, optimization strategy, adaptations to the mutual information measure, and a
progressive registration procedure. In addition to the theoretical method, the paper presents a experimental analysis
based on the quality of fit of final alignment between the LIDAR and digital imagery.