In recent years, many researches focused on the supervised classification, one of the machine learning methods, using Lidar and remotely sensed image to provide the four buildings, trees, roads, and grass categories of the ground features. However, buildings and trees are usually much closed or overlapped and this problem will lead buildings and nearby trees not easy to classify by single classification approach. The derived building outlines have many cracks which are not satisfactory for the requirement of GIS building vector map or building 3D modeling. To provide complete building outlines, this study develops an ¡§automatic detection of the overlapped areas of buildings and trees (ADOABT)¡¨ algorithm and an ¡§automatic linear feature recovery (ALFR)¡¨ approach to connect building outlines consequently. First, this research integrates Maximum Likelihood Classification (MLC) and Knowledge-Based Correction (KBC) to derive buildings and trees classification resultant images. Next, the ADOABT based on ¡§divide and conquer¡¨ principle was used to detect the overlapped areas of buildings and trees. Meanwhile, the building and tree edge images were detected using the Canny edge detector based on Lidar height image. Then, the intersection operator was applied to the detected areas and edge images to detect the crack of the building images. Afterward, vectorization and generalization of the intersection resultant images are applied to extract the straight line of the buildings. Finally, the automatic linear feature recovery procedure was performed to compensate the damage straight line effectively. According to the experiment results, the classification accuracy derived from integrated MLC and KBC classification method and the object-based classification (OBC) are similar. However, when applying the classification results to detect the overlapped areas of building and trees, because MLC and KBC has the procedure for handling temporal inconsistencies, the success rate of automatic detection is totally the same by artificial interpretation; the detection rate for the results of MLC and KBC is 100% whereas the one for the OBC only 67.7%. It can be concluded that the MLC and KBC approach is more suitable for the automatic detection for the overlapped areas of building and trees. Moreover, the ADOABT algorithm simplifies the workflow of the overlapped area detection. According to the result of edge detection and line detection, the Canny detector presents the clearest edge image. The lines extracted by Vectorized and generalization method are superior to the ones derived from Hough transform. The ALFR algorithm offers a way to connect building outline completely.

Multi-spectral Satellite imagery, among remotely sensed data from airborne and spaceborne platforms, contained the NIR band information is the major source for the land- cover classification. The main purpose of aerial imagery is for thematic land-use/land-cover mapping which is rarely used for land cover classification. Recently, the newly developed digital aerial cameras containing NIR band with up to 10cm ultra high resolution makes the land-cover classification using aerial imagery possible. However, because the urban ground objects are so complex, multi-spectral imagery is still not sufficient for urban classification. Problems include the difficulty in discriminating between trees and grass, the misclassification of buildings due to diverse roof compositions and shadow effects, and the misclassification of cars on roads. Recently, aerial LiDAR (ULiUght UDUetection UAUnd URUanging) data have been integrated with remotely sensed data to obtain better classification results. The LiDAR-derived normalized digital surface models (nDSMs) calculated by subtracting digital elevation models (DEMs) from digital surface models (DSMs) becomes an important factor for urban classification. This study proposed an adaptive raw-data-based, surface-based LiDAR data-filtering algorithm to generate DEMs as the foundation of generating the nDSMs. According to the experiment results, the proposed adaptive LiDAR data-filtering algorithm not only successfully filters out ground objects in urban, forest, and mixed land cover areas but also derives DEMs within the LiDAR data measuring accuracy based on the absolute and relative accuracy evaluation experiments results. For the aerial imagery urban classification, this study first conducted maximum likelihood classification (MLC) experiments to identify features suitable for urban classification using LiDAR data and aerial imagery. The addition of LiDAR height data improved the overall accuracy by up to 28 and 18%, respectively, compared to cases with only red¡Vgreen¡Vblue (RGB) and multi-spectral imagery. It concludes that the urban classification is highly dependent on LiDAR height rather than on NIR imagery. To further improve classification, this study proposes a knowledge-based classification system (KBCS) that includes a three-level height, ¡§asphalt road, vegetation, and non-vegetation¡¨ (A¡VV¡VN) classification model, rule-based scheme and knowledge-based correction (KBC). The proposed KBCS improved overall accuracy by 12 and 7% compared to maximum likelihood and object-based classification, respectively. The classification results have superior visual interpretability compared to the MLC classified image. Moreover, the visual details in the KBCS are superior to those of the OBC without involving a selection procedure for optimal segmentation parameters.