"The 5th International 3D GeoInfo Conference 2010 aims at bringing together international state-of-the-art research and facilitating the dialogue on emerging topics in the field of 3D geo-information. In recent years, research has focused on improving 3D data acquisition technology through remote sensing, photogrammetry, laser altimetry techniques (LiDAR), and visualization technology (3D CAD and Virtual Reality (VR) system) for 3D urban environments...."
source and more information: http://www.igg.tu-berlin.de/3dgeoinfo/
call for papers: http://www.igg.tu-berlin.de/3dgeoinfo/cfp.pdf
Showing posts with label ISPRS. Show all posts
Showing posts with label ISPRS. Show all posts
Tuesday, February 23, 2010
Thursday, September 24, 2009
A LiDAR Odyssey in Earth Observation
ISPRS Laserscanning `09 invited talk is online.
from
Very interesting and enjoyable pdf about Earth`s atmospheric LiDAR research.
http://laserscanning2009.ign.fr/download/LS09_Invited_talk_PHF.pdf
from
"Pierre H. Flamant
Laboratoire de Météorologie Dynamique,
Institut Pierre Simon Laplace
École Polytechnique"
Very interesting and enjoyable pdf about Earth`s atmospheric LiDAR research.
http://laserscanning2009.ign.fr/download/LS09_Invited_talk_PHF.pdf
Saturday, September 19, 2009
Full-Waveform filtering
Next generation scanners are integrated with full-waveform data acquisitions properties. Every single return pulse is documented. We can research the vertical conditions of the forest areas. Tree, bush, grass...different data class.
Interesting paper about full-waveform datasets....
INTEGRATION OF FULL-WAVEFORM INFORMATION INTO
THE AIRBORNE LASER SCANNING DATA FILTERING PROCESS
Y. -C. Lin and J. P. Mills
"ABSTRACT
Terrain classification of current discrete airborne laser scanning data requires filtering algorithms based on the spatial relationship between neighbouring three-dimensional points. However, difficulties commonly occur with low vegetation on steep slopes and when abrupt changes take place in the terrain. This paper reports on the integration of additional information from latest generation full-waveform data into a filtering algorithm in order to achieve improved digital terrain model (DTM) creation. Prior to a filtering procedure, each point was given an attribute based on pulse width information. A novel routine was then used to integrate pulse width information into the progressive densification filter developed by Axelsson. The performance was investigated in two areas that were found to be problematic when applying typical filtering algorithms. The derived DTM was found to be up to 0.7 m more accurate than the conventional filtering approach. Moreover, compared to typical filtering algorithms, dense low vegetation points could be removed more effectively. Overall, it is recommended that integrating waveform information can provide a solution for areas where typical filtering algorithms cannot perform well. Full-waveform systems are relatively cost-effective in terms of providing additional information without the need to fuse data from other sensors.
CONCLUSIONS
This study set out to investigate whether information derived from the latest generation full-waveform, small-footprint airborne laser scanning data could improve digital terrain modelling. A novel routine was designed to integrate waveform information into a commonly used filtering algorithm. The preliminary results have demonstrated that integrating pulse width information can provide a solution for areas where conventional filtering algorithms cannot perform well. On the top of an artificial mound, points rejected (Type I error) by a typical filtering algorithm can be correctly included in the developed routine. More low vegetation can also be correctly removed. However, rough or steep terrains with low vegetation cover, as well as forest terrain, still require further investigation and detailed validation. In addition, although identifying vegetation points becomes easier with the help of waveform information, it may be the case that in some densely vegetated areas insufficient “true” ground points exist for high-resolution DTM generation. In such cases, it might still be better to assume that the lowest point within a specified or adaptive window size is a ground point. The performance of existing filtering algorithms depends on the type of landscape (Sithole and Vosselman, 2004). Such approaches may require that users determine which type of landscape is being processed and then specify optimal parameters. As demonstrated in this paper, by using waveform information it is possible to automatically determine the landscape characteristics and then use the optimal parameter set for that specific landscape type. This will improve the automation of filtering procedures since less effort is required by users. Moreover, compared to ALS intensity values, pulse width information is easier to apply to different surveys since neither prior calibration or normalization procedures are required. Using full-waveform ALS data provides valuable physical and geometric information simultaneously. Such systems are relatively cost-effective in terms of providing multiple-information without the need to fuse data from other sensors."
Interesting paper about full-waveform datasets....
INTEGRATION OF FULL-WAVEFORM INFORMATION INTO
THE AIRBORNE LASER SCANNING DATA FILTERING PROCESS
Y. -C. Lin and J. P. Mills
"ABSTRACT
Terrain classification of current discrete airborne laser scanning data requires filtering algorithms based on the spatial relationship between neighbouring three-dimensional points. However, difficulties commonly occur with low vegetation on steep slopes and when abrupt changes take place in the terrain. This paper reports on the integration of additional information from latest generation full-waveform data into a filtering algorithm in order to achieve improved digital terrain model (DTM) creation. Prior to a filtering procedure, each point was given an attribute based on pulse width information. A novel routine was then used to integrate pulse width information into the progressive densification filter developed by Axelsson. The performance was investigated in two areas that were found to be problematic when applying typical filtering algorithms. The derived DTM was found to be up to 0.7 m more accurate than the conventional filtering approach. Moreover, compared to typical filtering algorithms, dense low vegetation points could be removed more effectively. Overall, it is recommended that integrating waveform information can provide a solution for areas where typical filtering algorithms cannot perform well. Full-waveform systems are relatively cost-effective in terms of providing additional information without the need to fuse data from other sensors.
CONCLUSIONS
This study set out to investigate whether information derived from the latest generation full-waveform, small-footprint airborne laser scanning data could improve digital terrain modelling. A novel routine was designed to integrate waveform information into a commonly used filtering algorithm. The preliminary results have demonstrated that integrating pulse width information can provide a solution for areas where conventional filtering algorithms cannot perform well. On the top of an artificial mound, points rejected (Type I error) by a typical filtering algorithm can be correctly included in the developed routine. More low vegetation can also be correctly removed. However, rough or steep terrains with low vegetation cover, as well as forest terrain, still require further investigation and detailed validation. In addition, although identifying vegetation points becomes easier with the help of waveform information, it may be the case that in some densely vegetated areas insufficient “true” ground points exist for high-resolution DTM generation. In such cases, it might still be better to assume that the lowest point within a specified or adaptive window size is a ground point. The performance of existing filtering algorithms depends on the type of landscape (Sithole and Vosselman, 2004). Such approaches may require that users determine which type of landscape is being processed and then specify optimal parameters. As demonstrated in this paper, by using waveform information it is possible to automatically determine the landscape characteristics and then use the optimal parameter set for that specific landscape type. This will improve the automation of filtering procedures since less effort is required by users. Moreover, compared to ALS intensity values, pulse width information is easier to apply to different surveys since neither prior calibration or normalization procedures are required. Using full-waveform ALS data provides valuable physical and geometric information simultaneously. Such systems are relatively cost-effective in terms of providing multiple-information without the need to fuse data from other sensors."
Pathway detection with LiDAR
Mapping of pathways is very difficult. We don't know exactly, where are the hidden ways in the forest, because we don't see through the trees on the aerial photos. :)....but, with LiDAR datasets that kind of mapping is possible.....important information is this, military reasons etc..
PATHWAY DETECTION AND GEOMETRICAL DESCRIPTION FROM ALS DATA IN
FORESTED MOUNTANEOUS AREA
Nicolas David, Cl´ement Mallet, Thomas Pons, Adrien Chauve, Fr´ed´eric Breta
"ABSTRACT:
In the last decade, airborne laser scanning (ALS) systems have become an alternative source for the acquisition of altimeter data. Compared to high resolution orthoimages, one of the main advantages of ALS is the ability of the laser beam to penetrate vegetation and reach the ground underneath. Therefore, 3D point clouds are essential data for computing Digital Terrain Models (DTM) in natural and vegetated areas. DTMs are a key product for many applications such as tree detection, flood modelling, archeology or road detection. Indeed, in forested areas, traditional image-based algorithms for road and pathway detection would partially fail due to their occlusion by the canopy cover. Thus, crucial information for forest management and fire prevention such as road width and slope would be misevaluated. This paper deals with road and pathway detection in a complex forested mountaneous area and with their geometrical parameter extraction using lidar data. Firstly, a three-step image-based methodology is proposed to detect road regions. Lidar feature orthoimages are first generated. Then, road seeds are both automatically and semi-automatically detected. And, a region growing algorithm is carried out to retrieve the full pathways from the seeds previously detected. Secondly, these pathways are vectorized using morphological tools, smoothed, and discretized. Finally, 1D sections within the lidar point cloud are successively generated for each point of the pathways to estimate more accurately road widths in 3D. We also retrieve a precise location of the pathway borders and centers, exported as vector data.
CONCLUSION
A full workflow for the pathway detection on mountainous area, from raw ALS data to vector database objects, have been proposed. With the increasing use of ALS data for DTM generation, such workflow should enabled to decrease the data acquisition cost for mapping institute. The detected pathways could also be used both for improving DTM generation and as features for strip adjustment and registration. The results show the feasibility of generating and updating pathway databases from ALS data, but their quality is still insufficient to be used on a production context for mapping agencies. In order to tackle the mentioned issues, it has been draw perspectives to improve robustness and automaticity of pathway detection."
PATHWAY DETECTION AND GEOMETRICAL DESCRIPTION FROM ALS DATA IN
FORESTED MOUNTANEOUS AREA
Nicolas David, Cl´ement Mallet, Thomas Pons, Adrien Chauve, Fr´ed´eric Breta
"ABSTRACT:
In the last decade, airborne laser scanning (ALS) systems have become an alternative source for the acquisition of altimeter data. Compared to high resolution orthoimages, one of the main advantages of ALS is the ability of the laser beam to penetrate vegetation and reach the ground underneath. Therefore, 3D point clouds are essential data for computing Digital Terrain Models (DTM) in natural and vegetated areas. DTMs are a key product for many applications such as tree detection, flood modelling, archeology or road detection. Indeed, in forested areas, traditional image-based algorithms for road and pathway detection would partially fail due to their occlusion by the canopy cover. Thus, crucial information for forest management and fire prevention such as road width and slope would be misevaluated. This paper deals with road and pathway detection in a complex forested mountaneous area and with their geometrical parameter extraction using lidar data. Firstly, a three-step image-based methodology is proposed to detect road regions. Lidar feature orthoimages are first generated. Then, road seeds are both automatically and semi-automatically detected. And, a region growing algorithm is carried out to retrieve the full pathways from the seeds previously detected. Secondly, these pathways are vectorized using morphological tools, smoothed, and discretized. Finally, 1D sections within the lidar point cloud are successively generated for each point of the pathways to estimate more accurately road widths in 3D. We also retrieve a precise location of the pathway borders and centers, exported as vector data.
CONCLUSION
A full workflow for the pathway detection on mountainous area, from raw ALS data to vector database objects, have been proposed. With the increasing use of ALS data for DTM generation, such workflow should enabled to decrease the data acquisition cost for mapping institute. The detected pathways could also be used both for improving DTM generation and as features for strip adjustment and registration. The results show the feasibility of generating and updating pathway databases from ALS data, but their quality is still insufficient to be used on a production context for mapping agencies. In order to tackle the mentioned issues, it has been draw perspectives to improve robustness and automaticity of pathway detection."
Friday, September 18, 2009
Geography and LiDAR II.
Interesting paper from Australia....
mapping of rivers from LiDAR datasets is very popular too....flood risk management.....
OPERATIONAL MAPPING OF THE ENVIRONMENTAL CONDITION OF RIPARIAN
ZONES OVER LARGE REGIONS FROM AIRBORNE LIDAR DATA
K. Johansen, L. Arroyo and S. Phinn
"ABSTRACT:
Riparian zones maintain water quality, support multiple geomorphic processes, contain significant biodiversity and also maintain the aesthetics of the landscape. Australian state and national government agencies responsible for managing riparian zones are planning missions for acquiring remotely sensed data covering the main streams in Victoria, New South Wales, and parts of Queensland and South Australia. The objectives of this paper are to: (1) assess the ability of LiDAR data for mapping the environmental condition of riparian zones; and (2) provide specifications for capturing and analyzing the LiDAR data for riparian zone mapping at large spatial extents (> 1000 km of stream length). LiDAR derived digital elevation models, terrain slope, intensity, fractional cover counts and canopy height models were used for mapping riparian condition indicators using simple algorithms and more complex objectoriented image analysis. The results showed that LiDAR data can be used to accurately map: water bodies (producer’s accuracy = 93%); streambed width (Root Mean Square Error (RMSE) = 3.3 m); bank-full width (RMSE = 6.1 m); riparian zone width (RMSE = 7.0 m); width of vegetation (RMSE = 5.6 m); plant projective cover (RMSE = 12%); vegetation height classes (vertical accuracy < r2 =" 0.40)."> 100,000 km of stream length."
mapping of rivers from LiDAR datasets is very popular too....flood risk management.....
OPERATIONAL MAPPING OF THE ENVIRONMENTAL CONDITION OF RIPARIANZONES OVER LARGE REGIONS FROM AIRBORNE LIDAR DATA
K. Johansen, L. Arroyo and S. Phinn
"ABSTRACT:
Riparian zones maintain water quality, support multiple geomorphic processes, contain significant biodiversity and also maintain the aesthetics of the landscape. Australian state and national government agencies responsible for managing riparian zones are planning missions for acquiring remotely sensed data covering the main streams in Victoria, New South Wales, and parts of Queensland and South Australia. The objectives of this paper are to: (1) assess the ability of LiDAR data for mapping the environmental condition of riparian zones; and (2) provide specifications for capturing and analyzing the LiDAR data for riparian zone mapping at large spatial extents (> 1000 km of stream length). LiDAR derived digital elevation models, terrain slope, intensity, fractional cover counts and canopy height models were used for mapping riparian condition indicators using simple algorithms and more complex objectoriented image analysis. The results showed that LiDAR data can be used to accurately map: water bodies (producer’s accuracy = 93%); streambed width (Root Mean Square Error (RMSE) = 3.3 m); bank-full width (RMSE = 6.1 m); riparian zone width (RMSE = 7.0 m); width of vegetation (RMSE = 5.6 m); plant projective cover (RMSE = 12%); vegetation height classes (vertical accuracy < r2 =" 0.40)."> 100,000 km of stream length."
Geography and LiDAR
researches at forest areas are a very popular part of the LiDAR know-how....
One of my favorite paper is from ISPRS Laserscanning '09.....very useful and innovative ideas, we need that kind of work, and not the discussions of RMS faults....how can I correct another mm-s....
A DECIDIOUS-CONIFEROUS SINGLE TREE CLASSIFICATION AND INTERNAL
STRUCTURE DERIVATION USING AIRBORNE LIDAR DATA
C. Ko, G. Sohn, T. K. Remmel
"ABSTRACT:
This project has two main purposes; the first is to perform deciduous-coniferous classification for 65 trees by using the leaf-on single flight LiDAR data. It was done by looking at the geometrical properties of the crown shapes (spherical, conical or cylindrical), these shapes were developed by a rule-driven method Lindenmayer Systems (L systems). Two more parameters that are data driven (convex hull analysis and buffer analysis) were developed to further capture the geometrical differences between deciduous and coniferous trees. Proposed methods are scale independent and arithmetically simple, they were developed simply by looking at the geometrical differences between the two types of trees. The classification rate was cross-validated and trees are 85% - 88% correctly classified. The second part of the project is to derive the internal structures of the LiDAR tree according to the results obtained from the first part. Internal structures include bole and branches; the location and orientation of the bole was done by connecting the geographic centres of horizontal slices of the tree. The branches were derived by k-means clustering techniques, different types of trees will yield a different type of branching structures for better visualization.
DISCUSSION / CONCLUSION
There are two major types of LiDAR systems for research and commercial use, full waveform and discrete returns. This paper has used only the discrete returns of the range data, and by studying the geometry of the crown shape properties, we classified the different crowns into two major classes, deciduous and coniferous. From the classified results, bole and branching structures were reconstructed according to the type of tree. The shape of the tree crown is inherited in the gene (adaptation) and therefore a certain species will have the similar crown shape and branching structures. The other factor affecting crown shape and branching structure is the growing strategies, which is adopted by the growing neighbour environment and those are more difficult to model (Horn, 1971). As a result, crown geometry is believed to be an important piece of information for species classification. By using just the three geometrical shapes (sphere, cone and cylinder), results were improved from 65% to 67% when the outliers were removed. If other parameters are included (area to volume ratio of convex hull and point to polygon buffering analysis), results were improved from 85% to 88%. Using crown shapes to classify trees is an intuitive method, but in this study it did not show promising results. By looking at the other geometrical properties, the results for classification increased considerably. Although different from what was expected, it is still believed crown shape and internal structure are good indicators for classifying trees, and future studies should be conducted in this direction. This method of classification is quite simple to produced and arithmetically easy. Tree bole and branches reconstructions are for visualization, but can also be used to study growth behaviour and to provide insight regarding why trees grow in a particular directions. These results are useful in many types of studies. For example, it can be used to study the potential hazards of a tree growing into structures, by classifying trees into deciduous and coniferous provide a better growth estimates."
One of my favorite paper is from ISPRS Laserscanning '09.....very useful and innovative ideas, we need that kind of work, and not the discussions of RMS faults....how can I correct another mm-s....
A DECIDIOUS-CONIFEROUS SINGLE TREE CLASSIFICATION AND INTERNAL
STRUCTURE DERIVATION USING AIRBORNE LIDAR DATA
C. Ko, G. Sohn, T. K. Remmel
"ABSTRACT:
This project has two main purposes; the first is to perform deciduous-coniferous classification for 65 trees by using the leaf-on single flight LiDAR data. It was done by looking at the geometrical properties of the crown shapes (spherical, conical or cylindrical), these shapes were developed by a rule-driven method Lindenmayer Systems (L systems). Two more parameters that are data driven (convex hull analysis and buffer analysis) were developed to further capture the geometrical differences between deciduous and coniferous trees. Proposed methods are scale independent and arithmetically simple, they were developed simply by looking at the geometrical differences between the two types of trees. The classification rate was cross-validated and trees are 85% - 88% correctly classified. The second part of the project is to derive the internal structures of the LiDAR tree according to the results obtained from the first part. Internal structures include bole and branches; the location and orientation of the bole was done by connecting the geographic centres of horizontal slices of the tree. The branches were derived by k-means clustering techniques, different types of trees will yield a different type of branching structures for better visualization.
DISCUSSION / CONCLUSION
There are two major types of LiDAR systems for research and commercial use, full waveform and discrete returns. This paper has used only the discrete returns of the range data, and by studying the geometry of the crown shape properties, we classified the different crowns into two major classes, deciduous and coniferous. From the classified results, bole and branching structures were reconstructed according to the type of tree. The shape of the tree crown is inherited in the gene (adaptation) and therefore a certain species will have the similar crown shape and branching structures. The other factor affecting crown shape and branching structure is the growing strategies, which is adopted by the growing neighbour environment and those are more difficult to model (Horn, 1971). As a result, crown geometry is believed to be an important piece of information for species classification. By using just the three geometrical shapes (sphere, cone and cylinder), results were improved from 65% to 67% when the outliers were removed. If other parameters are included (area to volume ratio of convex hull and point to polygon buffering analysis), results were improved from 85% to 88%. Using crown shapes to classify trees is an intuitive method, but in this study it did not show promising results. By looking at the other geometrical properties, the results for classification increased considerably. Although different from what was expected, it is still believed crown shape and internal structure are good indicators for classifying trees, and future studies should be conducted in this direction. This method of classification is quite simple to produced and arithmetically easy. Tree bole and branches reconstructions are for visualization, but can also be used to study growth behaviour and to provide insight regarding why trees grow in a particular directions. These results are useful in many types of studies. For example, it can be used to study the potential hazards of a tree growing into structures, by classifying trees into deciduous and coniferous provide a better growth estimates."
Outcrop Modeling
I like geography...our "terra" is so complex system, I think we shold be happy, that we here live....I open a beer....:)
Second paper about photogrammetry and LiDAR:
TERRESTRIAL LASER SCANNING COMBINED WITH PHOTOGRAMMETRY FOR
DIGITAL OUTCROP MODELLING
S. J. Buckley, E. Schwarz, V. Terlaky, J. A. Howell, R. W. C. Arnott
"ABSTRACT:
The integration of 3D modelling techniques is often advantageous for obtaining the most complete and useful object coverage for many application areas. In this paper, terrestrial laser scanning and digital photogrammetry were combined for the purposes of modelling a geological outcrop at Castle Creek, British Columbia, Canada. The outcrop, covering approximately 2.5 km2, comprised a smooth, scoured surface where recent glacial retreat had left the underlying sedimentary rocks exposed. The outcrop was of geological interest as an analogue to existing hydrocarbon reservoirs, and detailed spatial data were required to be able to map stratigraphic surfaces in 3D over the extent of the exposure. Aerial photogrammetry was used to provide a 2.5D digital elevation model of the overall outcrop surface. However, because the sedimentary strata were vertically orientated, local vertical cliffs acted as cross-sections through the geology, and these were surveyed using a terrestrial laser scanner and calibrated digital camera. Digital elevation models (DEMs) created from both methods were registered and merged, with the fused model showing a higher fidelity to the true topographic surface than either input technique. The final model was texture mapped using both the aerial and terrestrial photographs, using a local triangle reassignment to ensure that the most suitable images were chosen for each facet. This photorealistic model formed the basis for digitising the geological surfaces in 3D and building up a full 3D geocellular volume using these surfaces as input constraints. Because of the high resolution and accuracy of the input datasets, and the efficacy of the merging method, it was possible to interpret and track subtle surface separations over the larger extents of the outcrop.
CONCLUSIONS
Terrestrial laser scanning and digital aerial photogrammetry were combined to create a digital elevation model of the Castle Creek outcrop, British Columbia, Canada. The integration of the two techniques proved to be essential to capture both the large outcrop surface and the near-vertical cliff sections which were essential for being able to recreate the 3D orientation of geological surfaces. Use of surface matching allowed the aerial photogrammetric DEM to be accurately registered, without the problems of collecting a conventional photocontrol point arrangement in a rugged and remote area. Texture mapping with aerial and terrestrial images resulted in a photorealistic model that could be used by geologists for interpretation, education and quantitative analysis. This model demonstrated the application of geomatics for geological outcrop analogue modelling, allowing the spatial accuracy and resolution to be enhanced. A geocellular volume was created from digitised features, which will be used by geologists to improve the geological understanding of the Castle Creek outcrop."
Second paper about photogrammetry and LiDAR:
TERRESTRIAL LASER SCANNING COMBINED WITH PHOTOGRAMMETRY FOR
DIGITAL OUTCROP MODELLING
S. J. Buckley, E. Schwarz, V. Terlaky, J. A. Howell, R. W. C. Arnott
"ABSTRACT:
The integration of 3D modelling techniques is often advantageous for obtaining the most complete and useful object coverage for many application areas. In this paper, terrestrial laser scanning and digital photogrammetry were combined for the purposes of modelling a geological outcrop at Castle Creek, British Columbia, Canada. The outcrop, covering approximately 2.5 km2, comprised a smooth, scoured surface where recent glacial retreat had left the underlying sedimentary rocks exposed. The outcrop was of geological interest as an analogue to existing hydrocarbon reservoirs, and detailed spatial data were required to be able to map stratigraphic surfaces in 3D over the extent of the exposure. Aerial photogrammetry was used to provide a 2.5D digital elevation model of the overall outcrop surface. However, because the sedimentary strata were vertically orientated, local vertical cliffs acted as cross-sections through the geology, and these were surveyed using a terrestrial laser scanner and calibrated digital camera. Digital elevation models (DEMs) created from both methods were registered and merged, with the fused model showing a higher fidelity to the true topographic surface than either input technique. The final model was texture mapped using both the aerial and terrestrial photographs, using a local triangle reassignment to ensure that the most suitable images were chosen for each facet. This photorealistic model formed the basis for digitising the geological surfaces in 3D and building up a full 3D geocellular volume using these surfaces as input constraints. Because of the high resolution and accuracy of the input datasets, and the efficacy of the merging method, it was possible to interpret and track subtle surface separations over the larger extents of the outcrop.
CONCLUSIONS
Terrestrial laser scanning and digital aerial photogrammetry were combined to create a digital elevation model of the Castle Creek outcrop, British Columbia, Canada. The integration of the two techniques proved to be essential to capture both the large outcrop surface and the near-vertical cliff sections which were essential for being able to recreate the 3D orientation of geological surfaces. Use of surface matching allowed the aerial photogrammetric DEM to be accurately registered, without the problems of collecting a conventional photocontrol point arrangement in a rugged and remote area. Texture mapping with aerial and terrestrial images resulted in a photorealistic model that could be used by geologists for interpretation, education and quantitative analysis. This model demonstrated the application of geomatics for geological outcrop analogue modelling, allowing the spatial accuracy and resolution to be enhanced. A geocellular volume was created from digitised features, which will be used by geologists to improve the geological understanding of the Castle Creek outcrop."
Integration Approach of Photogrammetric and LiDAR
There are a lot of discussions, photogrammetric or LiDAR....I think, for this question difficult to find the answers, different projects, different solutions....maybe photogrammetric and LiDAR...we must use every opportunity to complete our work...
Here is a paper about integration techniques:
NEW INTEGRATION APPROACH OF PHOTOGRAMMETRIC AND LIDAR
TECHNIQUES FOR ARCHITECTURAL SURVEYS
F. Nex, F. Rinaudo
"ABSTRACT:
In the last few years, LIDAR and image-matching techniques have been employed in many application fields because of their quickness in point cloud generation. Nevertheless, these techniques do not assure complete and reliable results, especially in complex applications such as architectural surveys: laser scanning techniques do not allow the correct position of object breaklines to be extracted while image matching results require an accurate editing and they are not acceptable for bad-textured images. For this reason several authors have already suggested overcoming of these problems through a combination of LIDAR and photogrammetric techniques. These works considers the integration as the possibility to share point clouds generated by these techniques; however, a complete and automatic integration, in order to achieve more complete information, has never been implemented. In this paper, a new integration approach is proposed. This integration is focused on the possibility of overcoming the problems of each technique. In this approach LIDAR and multi-image matching techniques combine and share information in order to extract building breaklines in the space, perform the point cloud segmentation and speed up the modelling process in an automatic way. This integration is still an ongoing process: the algorithm workflow and first performed tests on real facades are presented in this paper, in order to evaluate the reliability of the proposed method; finally, an overview on the future developments is offered.
CONCLUSIONS AND FUTURE DEVELOPMENTS
The performed tests have allowed a first evaluation to be made of the potentiality of the proposed method, even though this analysis is not complete yet and further tests and changes have to be defined. Nevertheless, the results have already shown the reliability of the algorithm. In general, the results depend on the image taking configuration: almost normal case images are weak in the matching of edges parallel to epipolar lines. This problem could be overcome by just using more than three images and an ad hoc taking geometry. Images acquired at different heights allow epipolar lines with different direction to be obtained (Figure 9); instead, convergent images (more than 20°) could be subject to problems due to affine deformations. The taking distance should be chosen according to the degree of detail requested in the survey: in general, a 15 m distance can be considered the maximum for architectural objects to be drawn at 1:50 scale. The algorithm has shown that it can achieve good results for repetitive patterns, particularly if more than three images are used. The number of mismatches is usually low and decreases as the number of images increases. Glass, however, must always be deleted from all the images, in order to avoid mismatches. Dense point clouds in the LIDAR acquisition are not strictly necessary during the matching process while they are necessary instead in the filtering of the geometric edges. In fact the algorithm has shown some problems because of the presence of shadows close to the geometric breaklines; a more dense point cloud could overcome this problem. Furthermore, rounded edges are difficult to model as it is difficult to identify the position of the breakline in the image: in this situation, the algorithm does not allow good results to be obtained. It is expected that the geometric precision in the edge matching will be increased by implementing a Least Square Matching (LSM). A first step will be to perform a Multi-Photo LSM of dominants points; then it is planned to carry out an LS B-Snake matching (Zhang, 2005) which could slightly improve the quality of extracted edges. A great advantage in the traditional point cloud segmentation will be obtained from the extracted edges. The segmentation will be guided by edges that define the boundaries of each façade portion and fix a constraint in the region growing algorithm. In this way it is expected to make the modelling procedures easier."
Here is a paper about integration techniques:
NEW INTEGRATION APPROACH OF PHOTOGRAMMETRIC AND LIDAR
TECHNIQUES FOR ARCHITECTURAL SURVEYS
F. Nex, F. Rinaudo
"ABSTRACT:
In the last few years, LIDAR and image-matching techniques have been employed in many application fields because of their quickness in point cloud generation. Nevertheless, these techniques do not assure complete and reliable results, especially in complex applications such as architectural surveys: laser scanning techniques do not allow the correct position of object breaklines to be extracted while image matching results require an accurate editing and they are not acceptable for bad-textured images. For this reason several authors have already suggested overcoming of these problems through a combination of LIDAR and photogrammetric techniques. These works considers the integration as the possibility to share point clouds generated by these techniques; however, a complete and automatic integration, in order to achieve more complete information, has never been implemented. In this paper, a new integration approach is proposed. This integration is focused on the possibility of overcoming the problems of each technique. In this approach LIDAR and multi-image matching techniques combine and share information in order to extract building breaklines in the space, perform the point cloud segmentation and speed up the modelling process in an automatic way. This integration is still an ongoing process: the algorithm workflow and first performed tests on real facades are presented in this paper, in order to evaluate the reliability of the proposed method; finally, an overview on the future developments is offered.
CONCLUSIONS AND FUTURE DEVELOPMENTS
The performed tests have allowed a first evaluation to be made of the potentiality of the proposed method, even though this analysis is not complete yet and further tests and changes have to be defined. Nevertheless, the results have already shown the reliability of the algorithm. In general, the results depend on the image taking configuration: almost normal case images are weak in the matching of edges parallel to epipolar lines. This problem could be overcome by just using more than three images and an ad hoc taking geometry. Images acquired at different heights allow epipolar lines with different direction to be obtained (Figure 9); instead, convergent images (more than 20°) could be subject to problems due to affine deformations. The taking distance should be chosen according to the degree of detail requested in the survey: in general, a 15 m distance can be considered the maximum for architectural objects to be drawn at 1:50 scale. The algorithm has shown that it can achieve good results for repetitive patterns, particularly if more than three images are used. The number of mismatches is usually low and decreases as the number of images increases. Glass, however, must always be deleted from all the images, in order to avoid mismatches. Dense point clouds in the LIDAR acquisition are not strictly necessary during the matching process while they are necessary instead in the filtering of the geometric edges. In fact the algorithm has shown some problems because of the presence of shadows close to the geometric breaklines; a more dense point cloud could overcome this problem. Furthermore, rounded edges are difficult to model as it is difficult to identify the position of the breakline in the image: in this situation, the algorithm does not allow good results to be obtained. It is expected that the geometric precision in the edge matching will be increased by implementing a Least Square Matching (LSM). A first step will be to perform a Multi-Photo LSM of dominants points; then it is planned to carry out an LS B-Snake matching (Zhang, 2005) which could slightly improve the quality of extracted edges. A great advantage in the traditional point cloud segmentation will be obtained from the extracted edges. The segmentation will be guided by edges that define the boundaries of each façade portion and fix a constraint in the region growing algorithm. In this way it is expected to make the modelling procedures easier."
Friday, September 4, 2009
ISPRS Workshop Laserscanning 2009 I.
Hallo everybody,
Wednesday finished the ISPRS Laserscanning 2009 in Paris.
Thank you, congress was very professional!
160 people were from 25 countries at the workshop. 31 presenters, extra 4 company presentations and 27 posters were in the program.
31 presenters were from 13 countries. Iran and Greece cancelled.
(Country, number of oral presentation)
Austria | 3 |
Canada | 3 |
Finland | 4 |
France | 4 |
Germany | 6 |
Greece | 1 |
Iran | 1 |
Italy | 2 |
Japan | 1 |
Netherlands | 3 |
Swiss | 1 |
Spain | 1 |
UK | 1 |
Two areas (forestry, urban) are very popular in the LiDAR researches, where the newest papers were presented on the week.
I think, the most interesting topics were:
Automatic classification and feature extraction
TLS calibration
Photogrammetry and LiDAR combination
Analysis and visualisation with Voxel
Usage of LiDAR Intensity information
I want to write more details on the next days about these topics and the papers.
Thursday, August 27, 2009
ISPRS Workshop Laserscanning 2009
Report from the conferance on the TerraFormatics...coming soon.
Terms of Reference
Submissions are invited in, but not limited to, the following areas:
* Information extraction from point clouds
* Registration of point clouds
* LiDAR system improvements
* Analysis of full waveform lidar data
* Data management systems
* Feature extraction and 3D modelling
* Extraction of forest parameters from airborne/terrestrial lidar data
* Sensor modelling, calibration and validation
* Range imaging
* Data fusion
* Classification of natural and urban areas
* Deformation measurement and metrology
Program:
http://laserscanning2009.ign.fr/download/prog_web_laserscanning09.pdf
Subscribe to:
Posts (Atom)