Surveillance Video Georeference Method Based on Real Scene Model with Geometry Priors
Round 1
Reviewer 1 Report
In this study, authors have presented a method for Georeferencing of video captured by the surveillance camera.
Comments/Suggestions
· Line number 31, do not start the sentence with “And”
· Line numbers 46-49, rephrase the complete sentence
· Line numbers 46-49, rephrase the complete sentence
· Related literature about automatic feature point detection in overlapped images has not been mentioned in the literature. A few of them are:
o Forero, M. G., Mambuscay, C. L., Monroy, M. F., Miranda, S. L., Méndez, D., Valencia, M. O., & Gomez Selvaraj, M. (2021). Comparative Analysis of Detectors and Feature Descriptors for Multispectral Image Matching in Rice Crops. Plants, 10(9), 1791. https://doi.org/10.3390/plants10091791
o Sharma, S. K., Jain, K., & Shukla, A. K. (2023). A Comparative Analysis of Feature Detectors and Descriptors for Image Stitching. Applied Sciences, 13(10), 6015. https://doi.org/10.3390/app13106015
o Mukherjee, D.; Jonathan Wu, Q.M.; Wang, G. A comparative experimental study of image feature detectors and descriptors. Mach. Vis. Appl. 2015, 26, 443–466.
· More discussion can be included.
· Section 4.1, experimental equipment can be included in a table
· How registration accuracy was calculated, not clear
· Conclusions heading is missing
Grammatical mistakes are there in many places
Author Response
Dear reviewer, we appreciate the insightful comments and valuable suggestions from you. We have revised the manuscript substantially and incorporated your suggestions in this revised version. Responses to your questions and suggestions are given as follows.
Surveillance Video Georeference Method Based on Real Scene Model with Geometry Priors
No. remotesensing-2501019
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The manuscript presents an accurate registration method based on misalignment calibration and least squares matching between real scene and surveillance camera images to address this issue. The experiment results show that, the proposed method in this paper can achieve the expected goals of accurately matching real scene and surveillance camera images and assigning georeference information. The novelty is new and applicable. The reviewer suggests its acceptance after a major revision. The following issues and hesitations should be addressed:
1) The structure of the paper lacks the conclusion of the fifth part.
2) What is the real scene in the thesis? Where does it come from, such as in Fig.3 and Fig.4? How to ensure its accuracy?
3) In Fig.6, how to project a real scene? How to ensure its projection accuracy?
4) In Fig.8, how to select ground control points? Do you need to get their location information? If so, how to get them? What is the position accuracy requirement? How to ensure their position accuracy?
5) Section 4.3 and 4.4 need to describe the experimental process more clearly, for example, what is the meaning of Number in Table 4 and Table 5?
6) In Table 6, why do some cases fail? What is the reason?
7) How to define the accuracy rate? Is there a relationship between accuracy rate and success rate? Why?
Author Response
Dear reviewer, we appreciate the insightful comments and valuable suggestions from you. We have revised the manuscript substantially and incorporated your suggestions in this revised version. Responses to your questions and suggestions are given as follows.
Surveillance Video Georeference Method Based on Real Scene Model with Geometry Priors
No. remotesensing-2501019
Author Response File: Author Response.pdf
Reviewer 3 Report
The endeavor to imbue surveillance videos with georeferenced information is both intriguing and commendable. This paper presents a solution that proves effective and displays a commendable level of innovation. The manuscript exhibits a well-structured format and proficient writing style. While I find the overall content satisfactory, I do have a few suggestions for improvement as outlined below:
1. Clarification is needed regarding the determination of the initial value for the principal distance in formula (7).
2. In addition to the stipulation that the Ground Control Points (GCP) cannot be coplanar, it would be valuable to elucidate any further prerequisites concerning the spatial distribution of GCP during camera calibration.
3. The paper exclusively addresses smaller camera pitch angles (45 degrees, 55 degrees, and 65 degrees), which might not encompass the full range of real-world scenarios where the camera pitch could be larger. To fortify the paper's claims, a broader array of experimental scenarios should be considered.
4. The presence of lengthy sentences that pose challenges to comprehension is evident throughout the manuscript.
Based on the aforementioned considerations, I propose a minor revision to address the highlighted points.
Minor editing of the English language is required. For example, some sentences are too long.
Author Response
Dear reviewer, we appreciate the insightful comments and valuable suggestions from you. We have revised the manuscript substantially and incorporated your suggestions in this revised version. Responses to your questions and suggestions are given as follows.
Surveillance Video Georeference Method Based on Real Scene Model with Geometry Priors
No. remotesensing-2501019
Author Response File: 
Author Response.docx
Round 2
Reviewer 1 Report
All the comments/suggestions have been incorporated by the authors
Reviewer 2 Report
Accept in present form
