Lidar Classification

Automatic 3D extraction of building roofs from remotely sensed data is important for many applications including city modelling. This paper proposes a new method for automatic 3D roof extraction through an effective integration of LIDAR (Light Detection And Ranging) data and multispectral or-thoimagery. Using the ground height from a DEM (Digital Elevation Model), the raw LIDAR points are separated into two groups. The first group contains the ground points that are exploited to constitute a 'ground mask'. The second group contains the non-ground points which are segmented using an innovative image line guided segmentation technique to extract the roof planes. The image lines are extracted from the grey-scale version of the orthoimage and then classified into several classes such as 'ground', 'tree', 'roof edge' and 'roof ridge' using the ground mask and colour and texture information from the orthoimagery. During segmentation of the non-ground LIDAR points, the lines from the latter two classes are used as baselines to locate the nearby LIDAR points of the neighbouring planes. For each plane a robust seed region is thereby defined using the nearby non-ground LIDAR points of a baseline and this region is iteratively grown to extract the complete roof plane. Finally, a newly proposed rule-based procedure is applied to remove planes constructed on trees. Experimental results show

Automatic extraction of building roofs from remote sensing data is important for many applications, including 3D city modeling. This paper proposes a new method for automatic segmentation of raw LIDAR (light detection and ranging) data. Using the ground height from a DEM (digital elevation model), the raw LIDAR points are separated into two groups. The first group contains the ground points that form a " building mask ". The second group contains non-ground points that are clustered using the building mask. A cluster of points usually represents an individual building or tree. During segmentation, the planar roof segments are extracted from each cluster of points and refined using rules, such as the coplanarity of points and their locality. Planes on trees are removed using information, such as area and point height difference. Experimental results on nine areas of six different data sets show that the proposed method can successfully remove vegetation and, so, offers a high success rate for building detection (about 90% correctness and completeness) and roof plane extraction (about 80% correctness and completeness), when LIDAR point density is as low as four points/m 2. Thus, the proposed method can be exploited in various applications.

Airborne LiDAR is a widely accepted tool for archaeological prospection. Over the last decade an archaeology-specific data processing workflow has been evolving, ranging from raw data acquisition and processing, point cloud processing and product derivation to archaeological interpretation, dissemination and archiving. Currently, though, there is no agreement on the specific steps or terminology. This workflow is an interpretative knowledge production process that must be documented as such to ensure the intellectual transparency and accountability required for evidence- based archaeological interpretation. However, this is rarely the case, and there are no accepted schemas, let alone standards, to do so. As a result, there is a risk that the data processing steps of the workflow will be accepted as a black box process and its results as “hard data”. The first step in documenting a scientific process is to define it. Therefore, this paper provides a critical review of existing archaeology-specific workflows for airborne LiDAR-derived topographic data processing, resulting in an 18-step workflow with consistent terminology. Its novelty and significance lies in the fact that the existing comprehensive studies are outdated and the newer ones focus on selected aspects of the workflow. Based on the updated workflow, a good practice example for its documentation is presented.
Keywords: archaeological prospection; airborne LiDAR; method; workflow; documentation

Abundance, size, and spatial distribution of standing dead trees (snags), are key indicators of forest biodiversity and ecosystem health. These metrics represent critical habitat components for various wildlife species of conservation concern, including the Black-backed Woodpecker (Picoides arcticus), which is strongly associated with recently burned conifer forest. We assessed the potential of Airborne Laser Scanning (ALS) to detect and characterize conifer snags and identify Black-backed Woodpecker habitat using previously derived empirical thresholds of conifer snag basal area. Over the footprint of the Rim Fire, a megafire that extended (~ 104,000 ha) through a heterogeneous mosaic of conifer forests, oak woodlands, and meadows in the Sierra Nevada mountains of California, we identified conifer snags and estimated their basal area from single-tree ALS-derived metrics using Gaussian processes in four major steps. First, individual trees were mapped using the Watershed Segmentation algorithm, resulting in 87% detection of trees with stem diameter larger than 30 cm. Second, the snag/live classification model identified snags with an overall accuracy of 91.8%, using the coefficient of variation of height and intensity together with maximum intensity and fractional cover as the most relevant metrics. Third, the conifer/hardwood snag classification model utilizing the maximum height, median height, minimum intensity, and area metrics separated snag forest types with an overall accuracy of 84.8%. Finally, a Gaussian process regression model reliably estimated conifer snag stem diameter (R2 = 0.81) using height and crown area, thus significantly outperforming regionally calibrated conifer-specific allometric equations. As a result, ~ 80% of the snag basal area have been mapped. Optimal and potential habitat for Black-backed Woodpecker comprise 53.7 km2 and 58.4 km2, respectively, representing 5.1 and 5.6% of the footprint of the Rim Fire. Our study illustrates the utility of high-density ALS data for characterizing recently burned forests, which, in conjunction with information about the habitat needs of particular snag-dependent wildlife species, can be used to assess habitat characteristics, and thus contribute greatly to forest management and biodiversity conservation.

Content description: The problem of separating buildings from trees has been investigated using a combination of color, texture and dimensional cues for improved building detection performance in complex scenes. Abstract: Effective separation of buildings from trees is a major challenge in image-based automatic building detection. This paper presents a three-step method for effective separation of buildings from trees using aerial imagery and LIDAR data. Firstly, it uses cues such as height to remove objects of low height such as bushes, and width to exclude trees with small horizontal coverage. The height threshold is also used to generate a ground mask where buildings are found to be more separable than in so-called normalized DSM. Secondly, image entropy and color information are jointly applied to remove easily distinguishable trees. Finally, an innovative rule-based procedure is employed using the edge orientation histogram from the imagery to eliminate false positive candidates. The improved building detection algorithm has been tested on different test areas and it is shown that the algorithm offers high building detection rate in complex scenes which are hilly and densely vegetated. 2

This paper presents the results of the first attempt to assess, identify and quantify the residual number of shell craters of World War I currently present in the Vezzena/ Luserna/ Lavarone Plateau, areas of Millegrobbe, Bisele and Cima Campo (TN, Italy). Historical sources report the existence of several thousand of artillery explosions: therefore, a field survey or a classic photointerpretation appeared as labor-intensive and highly time-consuming. For this reason, a Digital Terrain Model (DTM) of the test-site was processed using the Sky-View Factor algorithm and was analyzed with an object-based approach, which implied: 1) multi-resolution segmentation; 2) classification (main features considered size, shape and color).
The automatically classified shell craters were thus verified during an in situ survey that determined the accuracy of the method in the order of 84% of the total occurrences.

The development of robust and accurate methods for automatic building detection and regularisation using multisource data continues to be a challenge due to point cloud sparsity, high spectral variability, urban objects differences, surrounding complexity, and data misalignment. To address these challenges, constraints on object's size, height, area, and orientation are generally benefited which adversely affect the detection performance. Often the buildings either small in size, under shadows or partly occluded are ousted during elimination of superfluous objects. To overcome the limitations, a methodology is developed to extract and regularise the buildings using features from point cloud and orthoimagery. The building delineation process is carried out by identifying the candidate building regions and segmenting them into grids. Vegetation elimination, building detection and extraction of their partially occluded parts are achieved by synthesising the point cloud and image data. Finally, the detected buildings are regularised by exploiting the image lines in the building regularisation process. Detection and regularisation processes have been evaluated using the ISPRS benchmark and four Australian data sets which differ in point density (1 to 29 points/m 2), building sizes, shadows, terrain, and vegetation. Results indicate that there is 83% to 93% per-area completeness with the correctness of above 95%, demonstrating the robustness of the approach. The absence of over-and many-to-many segmentation errors in the ISPRS data set indicate that the technique has higher per-object accuracy. While compared with six existing similar methods, the proposed detection and regularisation approach performs significantly better on more complex data sets (Australian) in contrast to the ISPRS benchmark, where it does better or equal to the counterparts.

This study scrutinises the use of terrestrial laser scanning (TLS) to measure diameter at breast height (DBH) and tree height at individual tree species level. LiDAR point cloud scans are collected from uniformly defined control points. The result of processed TLS data demonstrates the precise measurements of tree height and DBH by comparing it with field data (DBH, tree height, tree species and location). The average tree height and DBH obtained through TLS measurements were 9.44 m and 43.30 cm, respectively. A linear equation between TLS derived parameters and field measured values were established, which gave the coefficient of determination (r2) of 0.79 and 0.96 for tree height and DBH, respectively. Further, these parameters were used to calculate above ground biomass (AGB) for individual tree species by considering a non-destructive approach. The total AGB and carbon stock from 80 different trees are computed to be 49.601 and 22.320 tonnes, respectively.

Lidar is changing the paradigm of terrain mapping and gaining
popularity in many applications such as floodplain mapping,
hydrology, geomorphology, forest inventory, urban planning, and landscape ecology. One of the major barriers for a wider application of lidar used to be the high cost of data acquisition. However, this problem has been greatly alleviated with the thrilling developments in hardware. The first commercial airborne lidar system was introduced just ten years ago (Flood, 2001). Now, the latest system is capable of transmitting 100,000 pulses per second from an altitude of up to 2km. The pulse repetition rate has reached a maximum of more than 150 kHz and has increased by about 10-fold within the last 5 years; correspondingly, the cost of data collection has decreased by about 10 times within the same time period. Nowadays users can obtain data with a density of >1 pulses per m2 for several hundred dollars per square mile. The dramatically decreasing cost of data collection encourages more and more users to embrace this innovative technology in their application and research. For example, North Carolina has collected statewide
lidar to help the Federal Emergency Management Agency (FEMA) update their digital flood insurance rate maps (Stoker et al., 2006). A wealth of free lidar data are also accessible to the public from the websites maintained by governmental agencies such as the U.S. Geological Survey (the Center for Lidar Information, Coordination and Knowledge: CLICK), National Oceanic and Atmospheric Administration (Coastal Service Center), and U.S. Army Corps of Engineers (the Joint Airborne Lidar Bathymetry Technical Center of Expertise: JALBTCX).
Although lidar data has become more affordable for average
users, how to effectively process the raw data and extract useful information remains a big challenge. Compared to image processing, lidar is appealing in many aspects. For example, the users do not have to worry about geometric, atmospheric, and radiometric corrections. However, lidar data have some characteristics that post new challenges. First of all, lidar is essentially a kind of vector data. Different from raster data, the spatial locations of laser points have to be explicitly stored, making the file size much larger than imagery given the same “nominal” spatial resolution. Second, how to extract useful information from these seemingly random points is a relatively new research topic. The generation of digital elevation models (bare earth) is the largest and fastest growing application of
lidar data (Stoker et al., 2006). However, the research on automating the production of bare earth is still in its infancy. To make this situation worse, until recently, researchers tended not to publish their methods (Zhang et al., 2003, Chen et al., 2007). Besides terrain mapping, there is an endless list of areas where lidar has a potential application but they have not been adequately explored. I have developed a software (dubbed Tiffs: Toolbox for Lidar Data Filtering and Forest Studies) for processing lidar data and extracting bare earth and forest structure information. I will discuss the challenges and needs for lidar data distribution, management, and processing. I hope this article can shed light on the topic, not only for other software developers, but also for data providers and end users of lidar.

The hydrologic response of a catchment is sensitive to the morphology of the drainage network. Dimensions of bigger channels are usually well known, however, geometrical data for man-made ditches is often missing as there are many and small. Aerial LiDAR data offers the possibility to extract these small geometrical features. Analysing the three-dimensional point clouds directly will maintain the highest degree of information. A longitudinal and cross-sectional buffer were used to extract the cross-sectional profile points from the LiDAR point cloud. The profile was represented by spline functions fitted through the minimum envelop of the extracted points. The cross-sectional ditch profiles were classified for the presence of water and vegetation based on the normalized difference water index and the spatial characteristics of the points along the profile. The normalized difference water index was created using the RGB and intensity data coupled to the LiDAR points. The mean vertical deviation of 0.14 m found between the extracted and reference cross sections could mainly be attributed to the occurrence of water and partly to vegetation on the banks. In contrast to the cross-sectional area, the extracted width was not influenced by the environment (coefficient of determination R² = 0.87). Water and vegetation influenced the extracted ditch characteristics, but the proposed method is still robust and therefore facilitates input data acquisition and improves accuracy of spatially explicit hydrological models.

—This paper presents a new segmentation technique to use LIDAR point cloud data for automatic extraction of building roof planes. The raw LIDAR points are first classified into two major groups: ground and non-ground points. The ground points are used to generate a 'building mask' in which the black areas represent the ground where there are no laser returns below a certain height. The non-ground points are segmented to extract the planar roof segments. First, the building mask is divided into small grid cells. The cells containing the black pixels are clustered such that each cluster represents an individual building or tree. Second, the non-ground points within a cluster are segmented based on their coplanarity and neighbourhood relations. Third, the planar segments are refined using a rule-based procedure that assigns the common points among the planar segments to the appropriate segments. Finally, another rule-based procedure is applied to remove tree planes which are generally small in size and randomly oriented. Experimental results on three Australian sites have shown that the proposed method offers high building detection and roof plane extraction rates.

In this research, we propose the use of end-to-end deep learning simulation approach for assisting the design of LiDAR. The results show that two million points per second rate is optimal for point cloud based intersection classification task. The detection range of up to 100 meters corresponds to optimal classification performance. The 10 degree of upper field of view and 10 degree of lower field of view is sufficient for intersection classification. A linear increase of classification accuracy from 10 to 70 channels is evident. The research bridges the gap of lower level LiDAR simulation and development and self-driving visual tasks and expected to find applications to improve self-driving performance and safety.

Periodic building change detection is important for many applications, including disaster management. Building map databases need to be updated based on detected changes so as to ensure their currency and usefulness. This paper first presents a graphical user interface (GUI) developed to support the creation of a building database from building footprints automatically extracted from LiDAR (light detection and ranging) point cloud data. An automatic building change detection technique by which buildings are automatically extracted from newly-available LiDAR point cloud data and compared to those within an existing building database is then presented. Buildings identified as totally new or demolished are directly added to the change detection output. However, for part-building demolition or extension, a connected component analysis algorithm is applied, and for each connected building component, the area, width and height are estimated in order to ascertain if it can be considered as a demolished or new building-part. Using the developed GUI, a user can quickly examine each suggested change and indicate his/her decision to update the database, with a minimum number of mouse clicks. In experimental tests, the proposed
change detection technique was found to produce almost no omission errors, and when compared to the number of reference building corners, it reduced the human interaction to 14% for initial building map generation and to 3% for map updating. Thus, the proposed approach can be exploited for enhanced automated building information updating within a topographic database.

Filtering is a crucial step in lidar data processing. The Edge-based Morphological (EM) filtering method proposed by Chen et al. (2007, Photogrammetric Engineering and Remote Sensing, 73, pp. 175–185) is fast and can be applied to different land use and land cover types. However, it requires a large number of
parameters. It is challenging for average users to tune these parameters without a good understanding of the algorithm. This study introduces a new method to identify buildings so that the total number of parameters to be tuned is reduced
from 7 to 2. Even with fewer parameters being tuned, it was found that the average filtering error slightly decreased compared to the original algorithm when tested with the benchmark dataset provided by the International Society
for Photogrammetry and Remote Sensing (ISPRS) Commission III/WG3. This is a useful contribution to the original algorithm given that it can achieve increased accuracy in a simpler way for users.

Building detection in complex scenes is a non-trivial exercise due to building shape variability, irregular terrain, shadows, and occlusion by highly dense vegetation. In this research, we present a graph based algorithm, which combines multispectral imagery and airborne LiDAR information to completely delineate the building boundaries in urban and densely vegetated area. In the first phase, LiDAR data is divided into two groups: ground and non-ground data, using ground height from a bare-earth DEM. A mask, known as the primary building mask, is generated from the non-ground LiDAR points where the black region represents the elevated area (buildings and trees), while the white region describes the ground (earth). The second phase begins with the process of Connected Component Analysis (CCA) where the number of objects present in the test scene are identified followed by initial boundary detection and labelling. Additionally, a graph from the connected components is generated, where each black pixel corresponds to a node. An edge of a unit distance is defined between a black pixel and a neighbouring black pixel, if any. An edge does not exist from a black pixel to a neighbouring white pixel, if any. This phenomenon produces a disconnected components graph, where each component represents a prospective building or a dense vegetation (a contiguous block of black pixels from the primary mask). In the third phase, a clustering process clusters the segmented lines, extracted from multispectral imagery, around the graph components, if possible. In the fourth step, NDVI, image entropy, and LiDAR data are utilised to discriminate between vegetation, buildings, and isolated building's occluded parts. Finally, the initially extracted building boundary is extended pixel-wise using NDVI, entropy, and LiDAR data to completely delineate the building and to maximise the boundary reach towards building edges. The proposed technique is evaluated using two Australian data sets: Aitkenvale and Hervey Bay, for object-based and pixel-based completeness, correctness, and quality. The proposed technique detects buildings larger than 50 m 2 and 10 m 2 in the Aitkenvale site with 100% and 91% accuracy, respectively, while in the Hervey Bay site it performs better with 100% accuracy for buildings larger than 10 m 2 in area.

This paper presents a new segmentation technique for LIDAR point cloud data for automatic extraction of building roof planes. Using the ground height from a DEM (Digital Elevation Model), the raw LIDAR points are separated into two groups: ground and non-ground points. The ground points are used to generate a 'building mask' in which the black areas represent the ground where there are no laser returns below a certain height. The non-ground points are segmented to extract the planar roof segments. First, the building mask is divided into small grid cells. The cells containing the black pixels are clustered such that each cluster represents an individual building or tree. Second, the non-ground points within a cluster are segmented based on their coplanarity and neighbourhood relations. Third, the planar segments are refined using a rule-based procedure that assigns the common points among the planar segments to the appropriate segments. Finally, another rule-based procedure is applied to remove tree planes which are small in size and randomly oriented. Experimental results on the Vaihingen data set show that the proposed method offers high building detection and roof plane extraction rates.

Building data are one of the important data types in a topographic database. Building change detection after a period of time is necessary for many applications, such as identification of informal settlements. Based on the detected changes, the database has to be updated to ensure its usefulness. This paper proposes an improved building detection technique, which is a prerequisite for many building change detection techniques. The improved technique examines the gap between neighbouring buildings in the building mask in order to avoid under segmentation errors. Then, a new building change detection technique from LIDAR point cloud data is proposed. Buildings which are totally new or demolished are directly added to the change detection output. However, for demolished or extended building parts, a connected component analysis algorithm is applied and for each connected component its area, width and height are estimated in order to ascertain if it can be considered as a demolished or new building part. Finally, a graphical user interface (GUI) has been developed to update detected changes to the existing building map. Experimental results show that the improved building detection technique can offer not only higher performance in terms of completeness and correctness, but also a lower number of under-segmentation errors as compared to its original counterpart. The proposed change detection technique produces no omission errors and thus it can be exploited for enhanced automated building information updating within a topographic database. Using the developed GUI, the user can quickly examine each suggested change and indicate his/her decision with a minimum number of mouse clicks.

Automatic 3D reconstruction of building roofs from remotely sensed data is important for many applications including city modeling. This paper proposes a new method for automatic 3D roof reconstruction through an effective integration of LIDAR data and multi-spectral imagery. Using the ground height from a DEM, the raw LIDAR points are separated into two groups. The first group contains the ground points that are exploited to constitute a 'ground mask'. The second group contains the non-ground points that are used to generate initial roof planes. The structural lines are extracted from the grey-scale version of the orthoimage and they are classified into several classes such as 'ground', 'tree', 'roof edge' and 'roof ridge' using the ground mask, the NDVI image (Normalised Difference Vegetation Index from the multi-band orthoimage) and the entropy image (from the grey-scale orthoimage). The lines from the later two classes are primarily used to fit initial planes to the neighbouring LIDAR points. Other image lines within the vicinity of an initial plane are selected to fit the boundary of the plane. Once the proper image lines are selected and others are discarded, the final plane is reconstructed using the selected lines. Experimental results show that the proposed method can handle irregular and large registration errors between the LIDAR data and orthoimagery.

In this paper, the height variation in LIDAR (Light Detection And Ranging) point cloud data and point density are analyzed to remove the false building detection in highly vegetation and hilly sites. In general, the LIDAR points in a tree area have higher height variations than those in a building area. Moreover, the density of points having similar height values is lower in a tree area than in a building area. The proposed method uses such information as an improvement to a current state-of-the-art building detection method. The qualitative and object-based quantitative analyzes have been performed to verify the effectiveness of the proposed building detection method as compared with a current method. The analysis shows that proposed building detection method successfully reduces false building detection (i.e. trees in high complex sites of Australia and Germany), and the average correctness and quality have been improved by 6.36% and 6.16% respectively.

Automatic 3D extraction of building roofs from remotely sensed data is important for many applications including city modeling. This paper proposes a new method for automatic 3D roof extraction through an effective integration of LIDAR (Light Detection And Ranging) data and multispectral or-thoimagery. Using the ground height from a DEM (Digital Elevation Model), the raw LIDAR points are separated into two groups. The first group contains the ground points that are exploited to constitute a 'ground mask'. The second group contains the non-ground points which are segmented using an innovative image line guided segmentation technique to extract the roof planes. The image lines extracted from the grey-scale version of the orthoimage are classified into several classes such as 'ground', 'tree', 'roof edge' and 'roof ridge' using the ground mask and colour and texture information from the orthoimagery. During roof plane extraction the lines from the later two classes are used to fit roof planes to the neighbouring non-ground LIDAR points. Finally, a new rule-based procedure is applied to remove planes constructed on trees. Experimental results show that the proposed method successfully removes vegetation and offers high extraction rates.

There are currently several automatic building extraction methods introduced in the literature, but none of them are capable to completely extract portions of a building that are below a pre-defined building minimum height threshold. This paper proposes a systematic method which analyzes the height differences between the extracted adjacent planes above and below the height threshold as well as the planes' connectivity, thereby, extracting all portions belonging to buildings more completely. In general, the height difference between the edges of the adjacent planes above and below the height threshold that belong to the same building is more uniform. In addition, the extracted planes below the height threshold that belong to a building and their adjacent ground planes also have a clear height difference. The proposed method incorporates such information to achieve better performance in building extraction. We have compared our proposed method to a current state-of-the-art building extraction method qualitatively and quantitatively. Our experimental results show that our proposed method successfully recovers portions of a building below the height threshold, thereby achieving relatively higher average completeness (an improvement of 1.14%) and quality (an improvement of 0.93%).

There is a lot of buzz regarding the automobile industry, and the dominant goal for car manufacturers is to enhance drivers' experiences. Engine optimization and alternative fuels have made progress toward their full potential; so the next step is creating self-driving vehicles. The first group to explore this idea was DARPA in the early 2000s. Since 2010, major technological advancements have been achieved. As crazy as it may sound, self-driving vehicles are being developed, and research funding has shifted toward this goal. The future is coming, and within the next decade, the first commercial prototypes are expected to be available for mass use. When discussing self-driving vehicles, the first image that comes to mind is a complex combination of technologies, sensors and actuators, complicated algorithms, and high-performing computer systems. In this article, we take a step back and provide a top-down approach to investigating the technology behind autonomous vehicles.

This paper presents a segmentation of LIDAR point cloud data for automatic extraction of building footprint. Using the ground height information from a DEM (Digital Elevation Model), the non-ground points (mainly buildings and trees) are separated from the ground points. Points on walls are removed from the set of non-ground points. The remaining non-ground points are then divided into clusters based on height and local neighbourhood. Planar roof segments are extracted from each cluster of points following a region-growing technique. Planes are initialised using coplanar points as seed points and then grown using plane compatibility tests. Once all the planar segments are extracted, a rule-based procedure is applied to remove tree planes which are small in size and randomly oriented. The neighbouring planes are then merged to obtain individual building boundaries, which are regularised based on a new feature-based technique. Corners and line-segments are extracted from each boundary and adjusted using the assumption that each short building side is parallel or perpendicular to one or more neighbouring long building sides. Experimental results on five Australian data sets show that the proposed method offers higher correctness rate in building footprint extraction than a state-of-the-art method.

—Contour-based corner detectors directly or indirectly estimate a significance measure (eg, curvature) on the points of a planar curve and select the curvature extrema points as corners. A number of promising contour-based corner detectors have recently been proposed. They mainly differ in how the curvature is estimated on each point of the given curve. As the curvature on a digital curve can only be approximated, it is important to estimate a curvature that remains stable against significant noises, for example, geometric transformations and compression, on the curve. Moreover, in many applications, for instance, in content-based image retrieval, a fast corner detector is a prerequisite. So, it is also a primary characteristic that how much time a corner detector takes for corner detection in a given image. In addition, different authors evaluated their detectors on different platforms using different evaluation systems. Evaluation systems that depend on human judgements and visual identification of corners are manual and too subjective. Application of a manual system on a large test database will be expensive. Therefore, it is important to evaluate the detectors on a common platform using an automatic evaluation system. This paper first reviews six most recent and highly performed corner detectors and analyse their theoretical running time. Then it uses an automatic evaluation system to analyse their performance. Both the robustness to noise and efficiency are estimated to rank the detectors.

Separation of buildings from trees is a major challenge in automatic building detection. In residential and hilly areas, buildings are often surrounded by dense vegetation. This paper presents a three-step method for effective separation of buildings from trees. Firstly, height and width thresholds are applied to LIDAR data for removing small bushes and trees with small horizontal coverage, respectively. The generation of the building mask, where each black region indicates a void area from which there are no laser returns below the height threshold, also helps in separation of buildings from the nearby trees. Then image entropy and colour information are applied together to remove trees exhibiting high texture. Finally, an innovative rule-based procedure is employed using the edge orientation histogram from the imagery to eliminate the remaining trees. Experimental results show that the algorithm offers high building detection rate in complex scenes which are hilly and densely vegetated.

—Building boundary identification is an essential prerequisite in building outline generation from point cloud data. In this problem, boundary edges that constitute the building boundary are identified. The existing solutions to the identification of boundary edges from the input point set have one or more of the following problems: ineffective in finding appropriate edges in a concave shape, incapable of determining a 'hole' or 'concavity' inside the shape separately, dependant on additional information such as the scan direction that may be unavailable, and incompetent in determining the boundary of a point set from the boundaries of two or more subsets of the point set. This paper proposes a new solution to the identification of building boundary by using the maximum point-to-point distance in the input data. It properly detects the boundary edges for any type of shape and separately recognises holes, if any, inside the shape. The unique feature of the proposed solution is that it can identify the boundary of a point set from the boundaries of two or more subsets of the point set. It does not require any additional information other than the input point set. Experimental results show that the proposed solution can preserve details along the building boundary and offer high area-based completeness and quality, even in low density input data.

Using Airborne laser scanning (or LiDAR) for archaeological large-scale prospection of topography is especially interesting in forested regions where very subtle earthen archaeological features lie beneath the canopy, as at Phnom Kulen Mountain Plateau in Cambodia. In this thesis, the Phnom Kulen LiDAR data from the Cambodian Archaeological Lidar Initiative are processed for terrain analysis and visualization of archaeological topography. For systematic interpretation of these data, suitable visualization is necessary to allow a human interpreter the identification and mapping of archaeological topography. The key to successfully using various approaches in visualization of terrain datasets is to know the influences, advantages and disadvantages of different methods and how they contribute to identification and mapping of different topographical feature types. A comparison of previous studies on visualization techniques from ALS data shows, that there is no single visualization method which outperforms the rest in all types of terrain and for all types of features. Results of these studies are used to build hypotheses about suitability of different visualizations for feature identification and mapping in the study area. These hypotheses are examined at four representative test areas which include mounds, ponds, embankments and channels. All these archaeological microreliefs are extremely subtle and therefore hard to identify. Thus, besides analyzing which visualizations can be used to potentially identify and map them, the author presents the results of two analytical assessments in the study area. A range of quintessential methods were compared, which are: Standard hillshade, combined hillshades using principal-component-analysis, sky-view-factor, openness, slope, local relief model, local dominance and curvature (Laplacian-of-gaussian). The first assessment is based on a comparison of gradients of different visualization methods and spatial analysis using different terrain properties. It is analyzed under which circumstances-like different topographical features and terrain properties-different methods provide the highest gradients and therefore best perception of feature identification. In the second assessment the visualizations are used separately for mapping features and comparing the results. It is examined which visualizations best allow for identification and exact delineation of different archaeological topography. For linear earthworks the feature length of the mapping is compared, while for areal earthworks the difference in area of two mapping strategies (tight vs. loose) is measured. Those analyses show that the assumed hypotheses are not always true and thus results from previous studies are not transferable to this case study. However, relief manipulation like local relief models and visualizations based on openness and sky-view-factor are proven most suitable for identification and mapping of archaeological topography in the study area, depending on local terrain properties and feature types.

This paper presents a new LiDAR segmentation technique for automatic extraction of building roofs. First, it uses a height threshold, based on the digital elevation model to divide the LiDAR point cloud into 'ground' and 'non-ground' points. Then starting from the maximum LiDAR height, and decreasing the height at each iteration, it looks for coplanar points to form planar roof segments. At each height level, it clusters the points based on the distance and finds straight lines using the points. The nearest coplanar point to the midpoint of each line is used as a seed and the plane is grown in a region growing fashion. Finally, a rule based procedure is followed to remove planar segments in trees. The experimental results show that the proposed technique offers a high building detection and roof plane extraction rates while compared to a recently proposed technique.