An Improved Crack Identification Method for Asphalt Concrete Pavement
Round 1
Reviewer 1 Report
Dear authors
Thanks for your efforts, the article is a special work and has positive points about experiments and analytical models, but some things need to be revised:
-Good mathematical and computational analyzes have been done, but no specific practical results have been shown in this article, please provide more practical examples of these mathematical tasks.
- The authors know that the image feature enhancement method has been used in different ways in previous research, but what is the innovation of the current article?
-In relation 17, check, what is the ">" symbol for?
-Intersection and segmentation of the image is an important step in the automatic detection and classification of the type and especially the severity of the damage, and it has an important application for automatic crack sealing. What steps have been used in this method for this step?
-Noise in pavement crack images is effectively removed by median filtering, the image histogram correction technique is a very practical and useful intersection approach, and an edge detection algorithm is an ideal method to identify pavement crack borders.
-The SHRP LTPP protocol first classifies the type of cracks according to their direction, location, and shape, and quantifies the severity and extent of cracks according to their characteristics, including width, length, and area. Add a little more explanation about this protocol.
-Different algorithms are being used in the field of automatic analysis and description of pavement failures and many algorithms are being developed and tested. Display the position of this proposed method compared to other methods and techniques in a table or graph.
-Pavement crack data is sometimes used as the only criterion for pavement quality determination in many pavement management systems. This is especially true for road management systems used by local governments and in urban areas where road friction and roughness measurements are difficult due to lack of access to equipment, high cost, or relative inefficiency, It can't be done, it applies.Mention the importance of this issue briefly in the article.
respectfully
It seems that with a re-examination of the text, possible errors in the text can be fixed and corrected.
Author Response
Dear reviewer
We have made a peer-to-peer reply to your review comments, please check the uploaded PDF file for details.
Best regards,
Nan Yang
Author Response File: 
Author Response.pdf
Reviewer 2 Report
1. Well presented paper.
2.Any types of cracks detected by this method???
3. This method applicable only visible cracks???
4.For long length of road, how this method is feasible to detect the crack.
5.Objective not mentioned in the paper.
6. Explained de-noise processing.
1. English language appropriate in nature
Author Response
Dear reviewer
We have made a point-to-point reply to your review comments, please check the uploaded PDF file for details
Best regards,
Nan Yang
Author Response File: Author Response.pdf
Reviewer 3 Report
This paper presents a new approach on image feature enhancement and crack identification of cracks in asphalt concrete pavements. And the proposed scheme outperforms the state of the arts and can accurately detect multiple targets in multiple sensing applications. But there are some problems, which must be solved before it is considered for publication.
1. Relevant research background needs to be supplemented in INTRODUCTION.
2. The experiment and result analysis section can demonstrate the effectiveness of the method by showing the results of segmentation and recognition for a variety of pavement situations.
3. In the article, it is stated that " the algorithm in this paper can only process part of the pavement images", please explain the specific analysis of it.
The author should polish the language to improve the overall quality of the article.
Author Response
Dear reviewer
We have made a point-to-point reply to your review comments, please check the uploaded PDF file for details.
Best regards,
Nan Yang
Author Response File: Author Response.pdf
Reviewer 4 Report
This article using the Neigh Shrink algorithm to denoise the acquired image. Crack characteristics are segmented and processed by watershed segmentation algorithm. Through crack direction recognition and crack parameter extraction, crack distribution direction, crack length, width and other parameters of asphalt concrete pavement are obtained, so as to achieve accurate recognition of asphalt concrete pavement cracks. This paper has certain research significance and innovation. The specific review comments of this paper are as follows：
1、The main research of this paper is to improve the crack identification method, rather than to put forward some new concepts or methods, it is suggested to modify the title of the article.
2、For Table 3, it is recommended to supplement the original values of crack length and width in the table to make it easier for the reader to see the value of "Identify maximum deviation".
3、The final conclusion is "the algorithm in this paper can only process part of the pavement images", which suggests the shortcomings of the analysis method (such as how to process the image when the gap is filled by rain after rainfall).
4、The theoretical derivation in this paper is full, but the case data is insufficient. It is suggested to add some data analysis to make the conclusion of the theoretical derivation more persuasive.
5、In summary, it is suggested to minor repairs.
Minor editing of English language required.
Author Response
Dear reviewer
We have made a point-to-point reply to your review comments, please check the uploaded PDF file for details.
Best regards,
Nan Yang
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
The authors are thanked for the corrections.
Reviewer 3 Report
It can be accepted.
