ID: 57427
Title: Point-to-plane registration of terrestrial laser scans
Author: Darion Grant, James Bethel, Melba Crawford
Editor: George Vosselman
Year: 2012
Publisher: Elsevier, Vol 72, August 2012
Source: Centre for Ecological Sciences
Reference: None
Subject: ISPRS Journal of Photogrammetry and Remote Sensing
Keywords: LIDAR, point cloud, terrestrial laser scanning, surface registration
Abstract: The registration of pairs of Terrestrial Laser Scanning data (TLS) is an integral precursor to 3D data analysis. Of specific interest in this research work is the class of approaches that is considered to be fine registration and which does not require any targets or tie points. This paper presents a pairwise find registration approach called P2P that is formulated using the General Least Squares adjustment model. Given some initial registration parameters, the porposed P2P approach utilizes the scanned points and estimated planar features of both scans, along with their stochastic properties. These quantities are used to determine the optimum registration parameters in the least squares sense. The proposed P2P approach was tested on both simulated and real TLS data, and experimental results showed it to be four times more accurate than the registration approach of Chen and Medioni (1991).
Location: TE 12, Biologicalsciences Building, IISc
Literature cited 1: None
Literature cited 2: None
ID: 57426
Title: Bi-scale analysis of multitemporal land cover fractions for wetland vegetation mapping
Author: Ryo Michishita, Zhiben Jiang, Peng Gong, Bing Xu
Editor: George Vosselman
Year: 2012
Publisher: Elsevier, Vol 72, August 2012
Source: Centre for Ecological Sciences
Reference: None
Subject: ISPRS Journal of Photogrammetry and Remote Sensing
Keywords: Ecosystem, environment, land cover, monitoring, change detection, multitemporal
Abstract: Land cover fractions (LCFs) derived through spectral mixture analysis are useful in understanding subpixel information. However, few studies have been conducted on the analysis of time-series LCFs. Although multi-scale comparisons of spectral index, hard classification, and land surface temperature images have received attention, rarely have these approaches been applied to LCFs. This study compared the LCFs derived through Multiple Endmember Spectral Mixture Analysis (MESMA) using the time-series Landsat Thematic Mapper (TM) adn Terra Moderate Resolution Imaging Spectroradiometer (MODIS) data acquired in the Poyang Lake area, China between 2004 and 2005. Specifically, we aimed to : (1) propose an approach for optimal endmember (EM) seleciton in time-series MESMA; (2) understand the trends in time-series LCFs derived from the TM and MODIS data; and (3) examine the trends in the correlation between the bi-scale LCFs derived ffrom the time-series TM and MODIS data. Our results indicated: (1) The EM spectra chosen according to the proposed hierarchical three-step approach (overall, seasonal, and individual) accurately modeled the both the TM and MODIS images; (2) green vegetation (GV) and NPV/soil/impervious surface (N/S/I) classes followed sine curve trends in the overall area, while the two water classes displayed the water level change pattern in the areas primarily covered with wetland vegetation; adn (3) GV, N/S/I, and bright water classes indicated a moderately high agreement between the TM and MODIS LCFs in the whole area (adjusted R2?0.6). However, low levels of correlations were found in the areas primarily dominated by wetland vegetation for all land cover classes.
Location: TE 12, Biologicalsciences Building, IISc
Literature cited 1: None
Literature cited 2: None
ID: 57425
Title: Bi-scale analysis of multitemporal land cover fractions for wetland vegetatio mapping
Author: Ryo Michishita, Zhiben Jiang, Peng Gong, Bing Xu
Editor: George Vosselman
Year: 2012
Publisher: Elsevier, Vol 72, August 2012
Source: Centre for Ecological Sciences
Reference: None
Subject: ISPRS Journal of Photogrammetry and Remote Sensing
Keywords: Ecosystem, environment, land cover, monitoring, change detection, multitemporal
Abstract: Land cover fractions (LCFs) derived through spectral mixture analysis are useful in understanding subpixel information. However, few studies have been conducted on the analysis of time-series LCFs. Although multi-scale comparisons of spectral index, hard classification, and land surface temperature images have received attention, rarely have these approaches been applied to LCFs. This study compared the LCFs derived through Multiple Endmember Spectral Mixture Analysis (MESMA) using the time-series Landsat Thematic Mapper (TM) adn Terra Moderate Resolution Imaging Spectroradiometer (MODIS) data acquired in the Poyang Lake area, China between 2004 and 2005. Specifically, we aimed to : (1) propose an approach for optimal endmember (EM) seleciton in time-series MESMA; (2) understand the trends in time-series LCFs derived from the TM and MODIS data; and (3) examine the trends in the correlation between the bi-scale LCFs derived ffrom the time-series TM and MODIS data. Our results indicated: (1) The EM spectra chosen according to the proposed hierarchical three-step approach (overall, seasonal, and individual) accurately modeled the both the TM and MODIS images; (2) green vegetation (GV) and NPV/soil/impervious surface (N/S/I) classes followed sine curve trends in the overall area, while the two water classes displayed the water level change pattern in the areas primarily covered with wetland vegetation; adn (3) GV, N/S/I, and bright water classes indicated a moderately high agreement between the TM and MODIS LCFs in the whole area (adjusted R2?0.6). However, low levels of correlations were found in the areas primarily dominated by wetland vegetation for all land cover classes.
Location: TE 12, Biologicalsciences Building, IISc
Literature cited 1: None
Literature cited 2: None
ID: 57424
Title: Scan problems in digital CORONA satellite images from USGS archives
Author: Wouter Gheyle, Jean Bourgeois, Rudi Goossens, and Karsten Jacobsen
Editor: Russell G Congalton
Year: 2012
Publisher: ASPRS, Vol 77, No 12, December 2011
Source: Centre for Ecological Sciences
Reference: None
Subject: Photogrammetric Engineering and Remote Sensing
Keywords: CORONA, Earth Resources Observation and Science Center (EROS), US Geological Survey (USGS)
Abstract: The scientific value and relevance of declassified CORONA satellite images has been affirmed by numerous research projects and publications. From 1996 on, duplicates of the CORONA film were available in all standard analog photographic products, including film negatives and photo prints. Since September 2004, the analog imagery is no longer available adn has been replaced by digital images produced by the US Geological Survey (USGS) Earth Resources Observation and Science Center (EROS) Scanning Department. This paper points out to heretofore undetected and not negligile proglem with the digital imagery. A calibration error in the Leica DSW700 photogrammetric scanner has created gaps between scan tiles. We analyzed the effects these errors have on resulting DSMs and checked the extent of the scanning problem. Part of the USGS archive, i.e. images ordered and scanned between September 2004 and November-December 2007, have comparable scan errors but are nevertheless archived and available for future orders.
Location: TE 12, Biologicalsciences Building, IISc
Literature cited 1: None
Literature cited 2: None
ID: 57423
Title: Development of a seamless topographic/bathymetric digital terrain model for Tampa Bay, Florida
Author: Stephen C Medeiros, Tarig Ali, Scott C Hagen, and John P Raiford
Editor: Russell G Congalton
Year: 2012
Publisher: ASPRS, Vol 77, No 12, December 2011
Source: Centre for Ecological Sciences
Reference: None
Subject: Photogrammetric Engineering and Remote Sensing
Keywords: Digital terrain model (DTM), tide
Abstract: This applications paper presents the methods used to create a seamless topobathy digital terrain model (DTM) at 50-foot resolution intended to support hurricane storm surge modeling in Tampa Bay, Florida. Lidar, bathymetry, and various breakline data were integrated using the Terrain Data Set structure in ArcGIS?. The use of the Terrain Data Set structure allowed for embedding large data sets (such as lidar points)and archiving them after DTM creation while maintaing topographic analysis capabilities. The bathymetric data, natively referenced to Mean Sea Level (MSL), were converted to North American Vertical DAatum of 1988 (NAVD88) using an inverse distance weighted average offset from the three nearest NOAA tidal benchmark stations; results this conversion were within 6.1 centimeters of those produced by NOAA VDatum software in a quality control test area. This methodology can therefore be used in coastal regions of other countries.
Location: TE 12, Biologicalsciences Building, IISc
Literature cited 1: None
Literature cited 2: None
ID: 57422
Title: Landsat-5 TM and Lidar fusion for sub-pixel Juniper tree cover estimates in a Western rangeland
Author: Temuulen Sankey and Nancy Glenn
Editor: Russell G Congalton
Year: 2012
Publisher: ASPRS, Vol 77, No 12, December 2011
Source: Centre for Ecological Sciences
Reference: None
Subject: Photogrammetric Engineering and Remote Sensing
Keywords: Linear spectral unmixing (LSU),Constrained Energy Minimization (CEM), Mixture Tuned Matched Filtering (MTMF) techniques
Abstract: Pinyon-juniper woodlands comprise the third most common land-cover type in the United States and have been documented to have drastically increased both in density and extent in recent decades. We explored Landsat-5 TM and Ligth Detection and Ranging (lidar) data, individually and fused together, for estimating sub-pixel juniper cover. Linear spectral unmixing (LSU), Constrained Energy Minimization (CEM), and Mixture Tuned Matched Filtering (MTMF) techniques were compared along with spectral-lidar fusion approaches. None of the Landsat -5 TM-derived estimates were significantly correlated with field-measured juniper cover (n=100), while lidar-derived estimates were strongly correlated (R2 = 0.74, p - value<0.001). Fusion of these estimates produced superior results to both classifications individually (R2 = 0.80, p-value <0.001). The MTMF technique performed best, while a multiple regression -based fusion was the best approach to combining the two data sources. Future studies can use the best sub-pixel classification and fusion approach to quantify changes in associated ecosystem properties such as carbon.
Location: TE 12, Biologicalsciences Building, IISc
Literature cited 1: None
Literature cited 2: None
ID: 57421
Title: An assessment of internal neural network parameters affecting image classification accuracy
Author: Libin Zhou and Xiaojun Yang
Editor: Russell G Congalton
Year: 2012
Publisher: ASPRS, Vol 77, No 12, December 2011
Source: Centre for Ecological Sciences
Reference: None
Subject: Photogrammetric Engineering and Remote Sensing
Keywords: MLP networks, Enhanced Thematic Mapper Plus (ETM+) image, Gaussian Maximum Likelihood (GML)
Abstract: Neural networks are attractive intelligence techniques increasingly being used to classify remote sensor imagery. However, their performance is contingent upon a wide range of algorithm and non-algorithm factors. Despite significant progresses being made over the past two decades, there is no consistent guidance that has been established to automate the use of neural networks in remote sensing. The purpose of this study was to assess several internal parameters affecting image classification accuracy by multi-layer-perceptron (ML) neural networks. The MLP networks have been considered as the most popular neural network architecture. We carefully configured and trained a set of neural network models with different internal parameter settings. Then, we used these models to classify an Enhanced Thematic Mapper Plus (ETM+) image into several major land cover categories, and the accuracy of each classified map was assessed. The results reveal that number of hidden layers, activated function, and training rate cansubstantially affect the classification accuracy and that a neural network with appropriate internal parameters can lead to a significant classification accuracy improvement for urban land covers when comparing to the outcome by the Gaussian Maximum Likelihood (GML) classifier. These findings can help design efficient neural network models for improved performance.
Location: TE 12, Biologicalsciences Building, IISc
Literature cited 1: None
Literature cited 2: None
ID: 57420
Title: Building feature extraction from airborne lidar data based on tensor voting algorithm
Author: Rey-Jer You and Bo-Cheng Lin
Editor: Russell G Congalton
Year: 2012
Publisher: ASPRS, Vol 77, No 12, December 2011
Source: Centre for Ecological Sciences
Reference: None
Subject: Photogrammetric Engineering and Remote Sensing
Keywords: Tensor field, roof patches,
Abstract: This study presents a novel approach based on the tensor voting framework for extracting building features from airborne lidar data. Geometric features of lidar points are represented by a tensor field in this paper. For the extraction of roof patches, a region-growing method with principal features is developed from the properties of eigenvalues and eigenvectors of the tensor field. Additionally, three new indicators for the strength of features are presented to reduce the effect of the number of points on feature identification, and a supervised method is proposed to determine the threshold of planar feature strength for the region-growing. The extraction of ridge and edge lines from the segmented roof patches is also discussed. Experiments based on airborne lidar data are described to demonstrate the effectiveness of the proposed method, with those the results compared with the PCA method.
Location: TE 12, Biologicalsciences Building, IISc
Literature cited 1: None
Literature cited 2: None
ID: 57419
Title: Bias compensation in a rigorous sensor model and rational function model for high-resolution satellite images
Author: Tee-Ann Teo
Editor: Russell G Congalton
Year: 2012
Publisher: ASPRS, Vol 77, No 12, December 2011
Source: Centre for Ecological Sciences
Reference: None
Subject: Photogrammetric Engineering and Remote Sensing
Keywords: rigorous sensor model (RSM), rational function model (RFM), rational polynomial coefficients (RPCs)
Abstract: This paper presents three bias-compensated models for the geometric correction of high-resolution satellite images. The proposed models include the bias-compensated rigorous sensor model (RSM) in the orbital space, the bias-compensated RSM in the image space, and the bias-compensated rational function model (RFM) in the image space. The RSM and RFM use the on-board data and sensor-oriented rational polynomial coefficients (RPCs) provided in imagery metadata, respectively. Test images include QuickBird, WorldView - 1, and WorldView-2 Basic images. Experimental results indicate that the bias-compensated RSM using the zero order polynomials function in the orbital space provides higher accuracy. A comparison of the bias-compensated RSM and RFM in the image space shows that there models behave similarly, and the maximum difference in root-mean -square error is less than 0.1m. These results show that all the proposed methods obtain accuracy of better than 1 pixel, except for the translation in the image space.
Location: TE 12, Biologicalsciences Building, IISc
Literature cited 1: None
Literature cited 2: None
ID: 57418
Title: Tree topology representation from TLS point clouds using depth-first search in Voxel space
Author: Anita Schilling`, Anja Schmidt, and Hans-Gerd Maas
Editor: Russell G Congalton
Year: 2012
Publisher: ASPRS, Vol 78, No4, April 2012
Source: Centre for Ecological Sciences
Reference: None
Subject: Photogrammetric Engineering and Remote Sensing
Keywords: Circular Hough Transform, errestrial laser scanner.
Abstract: For a fundamental understanding of environmental processes and for the management of forests, information on the tree structure, preferably in 3D, is vital.Therefore, we propose a method to retrieve the spatial tree structure from 3D point clouds captured by a terrestrial laser scanner. The procedure addresses dense and noisy data sets of separate trees. Our method involves a variation of the Circular Hough Transform to determine trunk positions and a sequence of operations in voxel space. The core of the approach is the depth-first search algorithm, known from graph theory, to actually recover the tree as a graph. Furthermore, we compare results obtained from the tree graph to reference measurements of forest inventory parameters. The computation time of our method for topology representation is low and the method provides a reasonably accurate approximation of the 3D tree structure.
Location: TE 12, Biologicalsciences Building, IISc
Literature cited 1: None
Literature cited 2: None
ID: 57417
Title: An efficient point cloud management method based on a 3D R-tree
Author: Jun Gong, Qing Zhu, Ruofel Zhong, Yeting Zhang, and Xiao Xie
Editor: Russell G Congalton
Year: 2012
Publisher: ASPRS, Vol 78, No4, April 2012
Source: Centre for Ecological Sciences
Reference: None
Subject: Photogrammetric Engineering and Remote Sensing
Keywords: LoD (level of detail), 3DOR-Tree
Abstract: Vehicle-borne laser-scanned point clouds have become increasingly important 3D data sources in fields such as digital city modeling and emergency response management. Aiming at reducing the technial bottlenecks of management and visualization of very large point cloud data sets, this paper proposes a new spatial organization method called , which integrates Octree and 3D R-Tree data structures. This method utilizes Octree ' s rapid convergence to generate R-Tree leaf nodes, which are inserted directlyinto the R-tree, thus avoiding time-consuming point-by-point insertion operations. Furthermore, this paper extends the R-Tree structure to support LoD (level of detail) models. Based on the extended structure, a practical data management method is presented. Finally, an adaptive control method for LoDS of point clouds is illustrated. Typical experimental results show that our method possesses quasi-real-time index construction speed, a good storage utilization rate, and efficient visualization performance.
Location: TE 12, Biologicalsciences Building, IISc
Literature cited 1: None
Literature cited 2: None
ID: 57416
Title: Terrestrial laser scanning for delineating in-stream boulders and quantifying habitat complexity measures
Author: Jonathan P Resop, Jessica L Kozarek, and W Cully Hession
Editor: Russell G Congalton
Year: 2012
Publisher: ASPRS, Vol 78, No4, April 2012
Source: Centre for Ecological Sciences
Reference: None
Subject: Photogrammetric Engineering and Remote Sensing
Keywords: Terrestrial laser scanning (TLS), digital elevation model (DEM)
Abstract: Accurate stream topography measurement is important for many ecological applications such as hydraulic modeling and habitat characterization. Habitat complexity measures are often made using visual approximations or total station (TS) surveying that can be subjective and have limited spatial resolution. Terrestrial laser scanning (TLS) can measure topography at a high resolution and accuracy. two methods, TS surveying and TLS, were compared for measuring complex topography in a boulder-dominated 100 m forested reach of the Staunton River in Shenandoah National Park, Virginia. The mean absolute difference between the two datasets was 0.11 m with 82.3 percent of the TS data within + 0.1 m of TLS. The TLS dataset was processed to remove vegetation and create a 2 cm digital elevation model (DEM). An algorithm was developed for dlineating rocks within the stream channel from the DEM. A common ecological metric based on the structural complexity of the stream, percent in -stream rock cover, was calculated from the TLS dataset, and the results were compared to estimates from traditional methods. This aplication illustrates the potential of TLS to quantify habitat complexity measures in an automated, unbiased manner.
Location: TE 12, Biologicalsciences Building, IISc
Literature cited 1: None
Literature cited 2: None
ID: 57415
Title: Assessment of available rangeland woody plant biomass with a terrestrial lidar system
Author: Nian-Wei Ku, Sorin C Popescu, R James Ansley, Humberto L Perotto-Baldivieso, and Anthony M Filippi
Editor: Russell G Congalton
Year: 2012
Publisher: ASPRS, Vol 78, No4, April 2012
Source: Centre for Ecological Sciences
Reference: None
Subject: Photogrammetric Engineering and Remote Sensing
Keywords: Regression models, biomass, woody plants
Abstract: Woody plant encroachment directly threatens the grass forage production and reduces the area of cultivated lands. The rate of woody plant encroachment is increasing rapidly on rangelands of the southwestern US. However, woody plants, such as mesquite, are a possible source of bioenergy feedstock found on semi-arid lands. This study developed algorithms for determining woody plant biomass on rangelands at plot-level with a terrestrial lidar system. Two processing methods were investigated for analyzing the lidar point cloud data, namely: (a) percentile height statistics, and (b) a height bin approach. Regression models were developed for variables obtained through each processing technique and were able to explain 81 percent and 77 percent of the variance associated with the aboveground biomass using two processing methods, respectively. The results of this study revealed that terrestrial lidar can be used to accurately and efficiently estimate the aboveground biomass of woody plants in a semi-arid enviornment at plot level.
Location: TE 12, Biologicalsciences Building, IISc
Literature cited 1: None
Literature cited 2: None
ID: 57414
Title: Calibration and kinematic analysis of the Velodyne HDL-64E S2 lidar sensor
Author: Craig Glennie
Editor: Russell G Congalton
Year: 2012
Publisher: ASPRS, Vol 78, No4, April 2012
Source: Centre for Ecological Sciences
Reference: None
Subject: Photogrammetric Engineering and Remote Sensing
Keywords: laser scanner,
Abstract: A proposed methodology for the simultaneous kinematic calibration and boresight adjustment of the Velodyne HDL-64E S2 scanning lidar system is presented and analyzed. The mathematical model for both the measurements of the HDL-64E S2 scanner and their georeferencing by a GNSS/INS system is presented and discussed. A planar feature based least squares adjustment approach is utilized in order to derive an aoptimal solution for the laser ' s internal calibration parameters and boresight offsets. The results of the adjustment on an actual kinematic dataset along with a detailed examination of the adjustment residuals are given. An approximately 30 percent improvement in the 3D planar misclosure residual RMSE was achieved by the proposed calibration, as compared to an estimation of the laser boresight parameters only. Results also suggest that there may still be some un-modeled distortions in the range measurements from the scanner. However, despite this, the overall precision of the adjusted laser scanner data appears to make it a viable choice for high accuracy mobile scanning applications.
Location: TE 12, Biologicalsciences Building, IISc
Literature cited 1: None
Literature cited 2: None
ID: 57413
Title: Automated extraction of road markings from mobile Lidar point clouds
Author: Bisheng Yang, Lina Fang, Qingquan Li and Jonathan Li
Editor: Russell G Congalton
Year: 2012
Publisher: ASPRS, Vol 78, No4, April 2012
Source: Centre for Ecological Sciences
Reference: None
Subject: Photogrammetric Engineering and Remote Sensing
Keywords: Mobile lidar point clouds, Mobile Mapper
Abstract: Among implicit information of laser points, the strength of reflection of laser points is closely related to the property of road materials which helps detecting road markings. This paper presents a novel approach to automatically extracting road markings from mobile lidar point clouds. An interpolation method is first used to generate a georeferenced feature image of the point cloud, which helps to isolate the points of road surfaces. Then, an algorithm is used to separate these points within a range according to their strength of reflection. The separated points are further segmented to remove non-road points based on height threshold. Finally, the outlines of road markings are extracted from the segmented points using the semantic knowledge of road markings. The results demonstrated that our road markings from lidar point clouds collected by a land - based mobile lidar system, an Optech Lynx Mobile Mapper.
Location: TE 12, Biologicalsciences Building, IISc
Literature cited 1: None
Literature cited 2: None