the Creative Commons Attribution 4.0 License.
the Creative Commons Attribution 4.0 License.
| 06 May 2021
Study on a Compatible Model Combining Point Cloud Model and Digital Elevation Model
Abstract. DEMs (Digital Elevation Model) are important data sources that describe the surface morphology, but they are not real 3D models and thus cannot meet the requirements for describing the land surface in 3D. LIDAR (Light Deteation and Ranging) point cloud data are true 3D data with high precision and a high density. Based on an analysis of the differences between DEM and point cloud data, including the corresponding acquisition methods, data structures and model construction methods, this paper proposes a 3D point set data model based on regular grid 2D data field for regional modeling. The feasibility of the model is tested through the upper and lower boundary modeling method. The experiments show that (1) the 3D point set data model based on regularly gridded 2D data field is compatible with complete DEM data and simplified point cloud data and has good applicability; (2) the newly built data model can be used in the true 3D modeling of simple surface entities with high efficiency when the amount of data is only doubled; and (3) the new data model can be generated by inputting DEM data and point cloud data and using a simplified algorithm to process the point cloud data in the same coordinate system. This approach has the potential for multiscale, including large-scale, and automatic output processing and has the potential to be widely generalized.
This preprint has been withdrawn.
-
Withdrawal notice
This preprint has been withdrawn.
-
Preprint
(520 KB)
Interactive discussion
Status: closed
-
RC1: 'Comment on gi-2021-10', Anonymous Referee #1, 04 Jun 2021
The paper presents a method for the representation of 3D data through the use of Digital Elevation Models. It discusses the basic of DEM description of natural terrain, and then introduces an extension of the digital representation of topographic data in a DEM to account for multiple subsurface interfaces, such as the ceiling and roof of a cave.
The paper is well written and organized, and the language is clear. The presentation of the method is clear and concise. The conclusions are supported by the data. What to me seems to require a more in-depth discussion is how the present work compares to previous solutions to the problem of 3D data representation. Also, the quality of the resulting 3D representation is discussed in qualitative terms, without providing some numerical measure of the fidelity of the DEM to the original point cloud, for example. Lastly, limitations of the proposed methods are not discussed: how would a concave shape be represented, for example?
A final remark concerns the list of references: very often, standards and methodologies are quoted indirectly by citing papers that presumably contain references to the original works that present their description. I think that an effort should be made to quote the original sources.
Citation: https://doi.org/10.5194/gi-2021-10-RC1 -
AC1: 'Reply on RC1', Wenbo Guo, 18 Jun 2021
Thank you for your rigorous comment.Those comments are valuable and very helpful. We have read the comments carefully and reply as follows:
1.Comment:Â How the present work compares to previous solutions to the problem of 3D data representation.
1.Reply: The current 3D representation methods can be classified into three categories: surface-based modeling method, body-based modeling method and the modeling method that based on the mixture of surface and body. Some of those models have high accuracy, but in practical application, it is often difficult to promote. In the vast majority of large-scale terrain rendering projects, data representation methods other than TIN and DEM format are rarely used, and the results stored are basically in DEM format. In this work, the multi-level model based on regular grid is used for 3D representation. Compared with the current 3D representation solution, it has strong advantages in processing efficiency and scalability.
2.Comment: Also, the quality of the resulting 3D representation is discussed in qualitative terms, without providing some numerical measure of the fidelity of the DEM to the original point cloud, for example.
2.Reply: The original point cloud conversion to DEM can be divided into two steps:‘Point Cloud to Surface’and‘Surface to DEM’. There are extensive relevant studies on both processes. At the same time, because this problem is seriously affected by the scale and other factors, so this article does not explore this part deeply, and we may try to explore this problem in the subsequent work.
3.Comment: Lastly, limitations of the proposed methods are not discussed: how would a concave shape be represented, for example?
3.Reply: In part 2.3 of the article, we explained how to model the underground cavity through the model shown in Figure 3. The same interpretation applies to concave shapes.Of course, the model does have some limitations, which we have discussed in part 3.
4.Comment:Â A final remark concerns the list of references: very often, standards and methodologies are quoted indirectly by citing papers that presumably contain references to the original works that present their description. I think that an effort should be made to quote the original sources.
4.Reply: Thank you very much for the question and we will improve this issue in subsequent amendments.
Thank you once again for your attention to our paper.
Citation: https://doi.org/10.5194/gi-2021-10-AC1
-
AC1: 'Reply on RC1', Wenbo Guo, 18 Jun 2021
Interactive discussion
Status: closed
-
RC1: 'Comment on gi-2021-10', Anonymous Referee #1, 04 Jun 2021
The paper presents a method for the representation of 3D data through the use of Digital Elevation Models. It discusses the basic of DEM description of natural terrain, and then introduces an extension of the digital representation of topographic data in a DEM to account for multiple subsurface interfaces, such as the ceiling and roof of a cave.
The paper is well written and organized, and the language is clear. The presentation of the method is clear and concise. The conclusions are supported by the data. What to me seems to require a more in-depth discussion is how the present work compares to previous solutions to the problem of 3D data representation. Also, the quality of the resulting 3D representation is discussed in qualitative terms, without providing some numerical measure of the fidelity of the DEM to the original point cloud, for example. Lastly, limitations of the proposed methods are not discussed: how would a concave shape be represented, for example?
A final remark concerns the list of references: very often, standards and methodologies are quoted indirectly by citing papers that presumably contain references to the original works that present their description. I think that an effort should be made to quote the original sources.
Citation: https://doi.org/10.5194/gi-2021-10-RC1 -
AC1: 'Reply on RC1', Wenbo Guo, 18 Jun 2021
Thank you for your rigorous comment.Those comments are valuable and very helpful. We have read the comments carefully and reply as follows:
1.Comment:Â How the present work compares to previous solutions to the problem of 3D data representation.
1.Reply: The current 3D representation methods can be classified into three categories: surface-based modeling method, body-based modeling method and the modeling method that based on the mixture of surface and body. Some of those models have high accuracy, but in practical application, it is often difficult to promote. In the vast majority of large-scale terrain rendering projects, data representation methods other than TIN and DEM format are rarely used, and the results stored are basically in DEM format. In this work, the multi-level model based on regular grid is used for 3D representation. Compared with the current 3D representation solution, it has strong advantages in processing efficiency and scalability.
2.Comment: Also, the quality of the resulting 3D representation is discussed in qualitative terms, without providing some numerical measure of the fidelity of the DEM to the original point cloud, for example.
2.Reply: The original point cloud conversion to DEM can be divided into two steps:‘Point Cloud to Surface’and‘Surface to DEM’. There are extensive relevant studies on both processes. At the same time, because this problem is seriously affected by the scale and other factors, so this article does not explore this part deeply, and we may try to explore this problem in the subsequent work.
3.Comment: Lastly, limitations of the proposed methods are not discussed: how would a concave shape be represented, for example?
3.Reply: In part 2.3 of the article, we explained how to model the underground cavity through the model shown in Figure 3. The same interpretation applies to concave shapes.Of course, the model does have some limitations, which we have discussed in part 3.
4.Comment:Â A final remark concerns the list of references: very often, standards and methodologies are quoted indirectly by citing papers that presumably contain references to the original works that present their description. I think that an effort should be made to quote the original sources.
4.Reply: Thank you very much for the question and we will improve this issue in subsequent amendments.
Thank you once again for your attention to our paper.
Citation: https://doi.org/10.5194/gi-2021-10-AC1
-
AC1: 'Reply on RC1', Wenbo Guo, 18 Jun 2021
Viewed
| HTML | XML | Total | BibTeX | EndNote | |
|---|---|---|---|---|---|
| 956 | 223 | 49 | 1,228 | 42 | 42 |
- HTML: 956
- PDF: 223
- XML: 49
- Total: 1,228
- BibTeX: 42
- EndNote: 42
Viewed (geographical distribution)
| Country | # | Views | % |
|---|
| Total: | 0 |
| HTML: | 0 |
| PDF: | 0 |
| XML: | 0 |
- 1