Extraction of Roof Feature Lines Based on Geometric Constraints from Airborne LiDAR Data

Round 1

Reviewer 1 Report

The paper presents a strategy for extracting geometric feature lines based on LiDAR data, sequentially tracking the roof contours, normal ridges, oblique ridges, and valley lines. After normalization, fine feature lines are obtained. The following issues need to be improved:

1. In the introduction and related research, please add the technical route of detecting buildings from RGB images and then obtaining feature lines. It would be beneficial to include relevant research or content related to building detection, such as "Building extraction from remote sensing images with sparse token transformers" and "Automatic building extraction from very high resolution satellite imagery using line segment detector".

2. Please provide additional descriptions, including the differences between extracting feature lines from the masks obtained by the building detection algorithm and extracting feature lines based on LiDAR data. Also, it would be helpful to conduct relevant experiments to verify these differences.

3. I would like to see quantitative data for this method and comparison results with other methods. Please ensure these are included in your revised paper.

n/a

Author Response

Dear reviewer,

Thank you for reviewing our manuscript and for the constructive comments, which greatly helped us to improve the manuscript. We have heavily revised our experiments. The manuscript was carefully revised and point-by-point response was listed in the attachment.

Yours sincerely,

Zhan Cai

Author Response File: Author Response.pdf

Reviewer 2 Report

The manuscript presents a new algorithm for delineating roof feature lines from LiDAR data.

The approach appears scientifically sound and the literature review adequate. The manuscript is very technical and generally well written. However, explanations are occasionally missing so that the reader can understand the algorithm.

The manuscript still needs some improvements. For example, it refers to six datasets to evaluate the algorithm, but only half of them can be used. Fortunately, the remaining half is more than enough to achieve the algorithm proof of concept. The manuscript also suggests several parameter values, but it is unclear whether these values can be rigorously justified or whether they result from empirical estimates based on the datasets used. Finally, the discussion should be improved to help the reader weigh the pros and cons of the proposed algorithm.

Detailed comments are provided in an annotated version of the manuscript uploaded with this report. Minor comments are in yellow, major comments are in red, and highlights are in green.

Comments for author File: Comments.pdf

I detected a few misspelled words and the use of the term “vertical” instead of "orthogonal".

Author Response

Dear reviewer,

Thank you for reviewing our manuscript and for the constructive comments, which greatly helped us to improve the manuscript. We have heavily revised our experiments. The manuscript was carefully revised and point-by-point response was listed below. The most comments have been responded in the revised manuscript. The remain comments are laid out below in italicized font and specific concerns have been numbered. Our response is given in normal font and changes/additions to the manuscript are given in blue text. We hope that your comments have been addressed accurately. The changes have been marked yellow in the revised manuscript, and the replies of each comment are presented in blue text as follows.

1. “...should be firstly projected to two-dimension XOY plane”, where “XOY” should be modified to “XY”?

Response: Thanks for your valuable advice. “XOY” may be a common expression in LiDAR data processing, where “O” represents the coordinate origin.

2. In section 3.1.1, you state that the first step in your algorithm is to roughly delineate roof outlines. Here you show that the first step is to identify edge points in the data (Figure 7), the second is to delineate roof outlines (Figure 8). Remove either Figure 7, or add a step in section 3.1.

Response: Thanks for your valuable advice. We would like to explain that, in “Coarse roof outlines” part, the third step is “detecting edge points”. Figure 7 (Figure 9 in the revised manuscript) shows the extraction effect of edge points, aiming at illustrating that the edge points obtained by the proposed method are enough for the subsequent extraction of roof outlines efficiently.

Thanks very much for the valuable advices again.

Yours sincerely,

Zhan Cai

Round 2

Reviewer 1 Report

The author have addressed my problems

n/a