A Hybrid Method to Incrementally Extract Road Networks Using Spatio-Temporal Trajectory Data
Round 1
Reviewer 1 Report
The paper describes an approach to extract road networks from trajectory data. The paper is technically feasible and presents an evaluation with imaging data including a comparison to another technique. I am not deep enough into the topic to judge whether the methods that the authors compare against is considered the state of the art. As the other methods is really producing bad results, I doubt that this really reflects the state of the art.
How well the presented methods really works is hard to say, as the overlay of trajectories over the image does not allow me to see the roads of the image. They are occluded by all the lines. For example, the gravitation force that is used could theoretically merge two close-by roads. From the results presented, it is not clear to me, whether this happened.
Also, I am not sure whether all heuristics for merging time slices are correct. For example, in Figure 7, L2 could be a dead-end street. Is it tested that whether trajectories actually continue from L2 to L3, L4, or L5?
Minor comments:
The term "multi-temporal" is weird. There is only one time, not multiple. What the authors mean are time slices. A better term should be used. How is k chosen in Equation (4)? The approach in Equation (4) is similar to a k-means procedure. It would be good to discuss this. There are many spelling mistake that a spell-checker would easily detect.Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 2 Report
The article presents an interesting methodology for obtaining the road graph from a multi-temporal surveys series.
The methodology description is good and the results clearly presented.
Author Response
Response to Reviewer 2 Comments
Point 1: The article presents an interesting methodology for obtaining the road graph from a multi-temporal survey series. The methodology description is good and the results clearly presented.
Response 1: Thank you very much for your positive comments. We also hope the paper can be published in IJGI as soon as possible.
Reviewer 3 Report
Interesting paper, still, in my opinion, highly penalized by the frequent English grammar error and typos.
For instance:
L. 27: remove "respectively"
L. 45: what is Ubiquitous positioning? a discipline? reference
L. 72: again: "on the other hand"
Typos:
L. 28: inconsistnecies
L. 104: temproal
L. 115 achievie
... and many more.
The paper, scientifically valid and supported by apparently good results (although the comparison with reference [32] is probably not exhaustive), needs a strong grammar and syntactic review.
Author Response
Response to Reviewer 3 Comments
Point 1: Interesting paper, still, in my opinion, highly penalized by the frequent English grammar error and typos. For instance:
L. 27: remove "respectively"
L. 45: what is Ubiquitous positioning? a discipline? reference
L. 72: again: "on the other hand"
Typos:
L. 28: inconsistnecies
L. 104: temproal
L. 115 achievie
... and many more.
Response 1: Thank you for your comments. We have revised these grammar errors and typos and sent the paper to an English editing service for language spell check. Particularly, the term “ubiquitous positioning” is changed to “Widely used positioning devices”.
Point 2: The paper, scientifically valid and supported by apparently good results (although the comparison with reference [32] is probably not exhaustive), needs a strong grammar and syntactic review.
Response 2: Thank you for your comments. In addition to sending the paper to MDPI for a professional English check, we also enriched the experimental analysis by qualitatively and quantitively comparing our method with another three methods of literatures [27,34,39].
Reviewer 4 Report
Dear Authors,
Very interesting experiment was conducted using a hybrid method for incrementally extracting urban road networks from spatio-temporal trajectory data. Really seems very nice effort for choosing an innovative idea in the enhance of approaches for road network data extraction and updating.
It worth to mention that the title of this paper sufficiently clearly reflect its contents. The subject suitable for publication in this journal. The key-words could be more informative. The abstract is informative, concise and complete. Very detailed review of the literature is made.
At the same time, I would like to draw the attention of the authors to the fact that the “conclusion” could be much stronger.
I would also advise to improve the quality and size of some figures (1, 6, 8, 9, 12).
Best regards.
Author Response
Response to Reviewer 4 Comments
Point 1: Very interesting experiment was conducted using a hybrid method for incrementally extracting urban road networks from spatio-temporal trajectory data. Really seems very nice effort for choosing an innovative idea in the enhance of approaches for road network data extraction and updating. It worth to mention that the title of this paper sufficiently clearly reflect its contents. The subject suitable for publication in this journal. The key-words could be more informative. The abstract is informative, concise and complete. Very detailed review of the literature is made. At the same time, I would like to draw the attention of the authors to the fact that the “conclusion” could be much stronger.
Response 1: Thank you very much for your positive comments. We have revised the key-words and rewritten the conclusion part for stronger informative description.
Point 2: I would also advise to improve the quality and size of some figures (1, 6, 8, 9, 12).
Response 2: Thank you very much for your kindly advise. We have modified Figure 1, 6, 8, 9 and 12 to ensure suitable picture quality and size.