Research on Urban Road Network Extraction Based on Web Map API Hierarchical Rasterization and Improved Thinning Algorithm

Round 1

Reviewer 1 Report

This manuscript shows the trajectory generation and road network extraction method based on map API proposed has a good extraction effect, effectively solving the road network acquisition problem in spatial planning. Also, it provides a new idea for fine-scale road network extraction research. The conclusions are solid and convincing, and the manuscript is of high quality. 

Author Response

According to the advice of academic editors, we have added content on the contribution of the sustainable impact of urban transport. And this manuscript has undergone English language editing by MDPI. The text has been checked for correct use of grammar and common technical terms, and edited to a level suitable for reporting research in a scholarly journal.

Reviewer 2 Report

The paper presents the method of urban road network extraction and resolves a long-standing issue of urban GPS trajectory data being large and difficult to obtain.

Overall, this paper provides a new perspective on the research of road network extraction compared with the existing literature. the paper is well organized and its presentation is fluent. However, some minor issues still need to be improved:

(1) The typesetting of individual image names needs to be adjusted; such as in Figure 3.

(2) I suggest the literature should be updated, and some important papers should be cited.

 

(3) Every research should have a future work so add it in the last of the conclusion section. Besides, the conclusion should be concise and only summarize the most significant contribution of the research.

(4) The authors have omitted some important references, which should be added in the manuscript. 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

The research work is interesting and well constructed. From my point of view, extracting the road network by this method has difficulties in obtaining the database with vehicle routes and their accuracy, especially in a highly urbanized area, like this one of the authors.

From a scientific point of view, I had difficulties in understanding:

1. the way in which each point of the work image is generated, an explicit mathematical formulation being necessary

2. how to extract the trajectory for the studied area if the route was larger than the studied area

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

The study is quite interesting. The topic is related to trajectory generation and urban road network extraction based on map API.

In my opinion, the main problem of this work is that it is not really clear how you achieved this goal: what data did you use? GPS trajectories? How do you get them? Is it possible to see what they look like? Are you really sure that this method is more accurate (see line 43) compared with traditional methods, or does it depend on the device accuracy? How the trajectories are related to the vector images? Please try to better explain which are the input data and how they are treated to obtain the results.

In addition, there is a cursory description of the literature review. Please increase your effort to frame the question.

Moreover, there are lots of statement to be explored, such as:

- line 53 "... relevant road information", which kind of information?

- line 63/64 50 mx50 m, why have you decided to take this distances?

- line 69 "... does not make sense", why?

- line 86 "... the trajectory too short", why?

- line 101 Table 1 road_types, what is it?

- line 105 "... are merged into one trajectory", how?

and so on. Please take nothing for granted.

Before continuing the review of the Results, Discussion, and Conclusion, it's a priority to completely reorganize your speech (Introduction and Materials and methods) to help readers in better understanding how is the work structured, which are the input data, how they are related to each other and how they are treated.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 4 Report

I would like to thank the Authors for the valuable changes in the paper, which is now clearer and more detailed. 

I would like to focus the attention on some other changes and explanations which should be made:

1. the abstract is too long; it is composed of 284 words, and the policy of the journal is to have abstracts with 200 words max. Please reduce it.

2. is there any problems with GPS data in tunnels or in places where the signal is too low?

3. have you got an homogeneous data sample with the Didi Chuxing GAIA source? 

4. which are the differences between the three different levels Div, Grid, and Cell? You can better explain that.

5. what does "can be added by one" mean at line 110?

6. there is a mistake at line 124 (i.e. longi4182tude).

7. at lines 100 and 102 the Figure is probably the second one.

8. moderate English changes are required (i.e. lines 108, 112-113, 117-120, and so on).

9. the reference bibliography shoul be improved a little more. There are 17/31 Chinese references (more than half); try to widen your search. Here are some suggestions:

- SONI, Sonam; KAUR, Sukhmeet. To Propose an improvement in Zhang-Suen algorithm for image thinning in image processing. Int. J. Sci. Technol. Eng., 2016, 3.1: 7481.

- CANTISANI, Giuseppe; DEL SERRONE, Giulia. Procedure for the identification of existing roads alignment from georeferenced points database. Infrastructures, 2020, 6.1: 2.

- CHEN, Daniel, et al. Road network reconstruction for organizing paths. In: Proceedings of the twenty-first annual ACM-SIAM symposium on Discrete Algorithms. Society for Industrial and Applied Mathematics, 2010. p. 1309-1320.

- DEY, Tamal K.; WANG, Jiayuan; WANG, Yusu. Road network reconstruction from satellite images with machine learning supported by topological methods. In: Proceedings of the 27th ACM SIGSPATIAL International Conference on Advances in Geographic Information Systems. 2019. p. 520-523.

- SONI, Pramod Kumar; RAJPAL, Navin; MEHTA, Rajesh. Road Centerline Extraction From VHR Images Using SVM and Multi-Scale Maximum Response Filter. Journal of the Indian Society of Remote Sensing, 2021, 49.7: 1519-1532.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf