Real-Time Detection of Nickel Plated Punched Steel Strip Parameters Based on Improved Circle Fitting Algorithm
Round 1
Reviewer 1 Report
All is as in the report.
Comments for author File: 
Comments.pdf
Author Response
I can recommend it for publication if the authors accept the following suggestions
- a) A_{x}, B_{x} instead as in text
- b) X_{1}, X_{2},…instead as in text
- c) Hence, check all the formules in the whole text;
- d) Check the names of all journal in the list of references including the abbreviated of oness.
Etc,…
Answer 1: Thank you for your kind comments. (Ax,Ay) and (Bx,By) have been revised to (Ax, Ay) and (Bx, By). See lines 178-181 for details.
Answer 2: Thank you for your kind comments. {X1,X2,...,Xn} has been revised to {X1,X2,...,Xn}, please see lines 166 and 188 of revised version.
Answer 3: Thank you for your kind comments. We have rechecked the full text and further standardized the description of the formula, especially the subscript of the formula, as shown in the red marked formula section of the revised version.
Answer 4: Thank you for your kind comments. We have checked the names of all journal in the list of references including the abbreviated of oness.
Thank you for all your valuable suggestions and advice, we are very grateful and will try our best to modify the paper until it meets all your requirements and expectations. Please let me know if any further revision is necessary.
Author Response File: Author Response.pdf
Reviewer 2 Report
Please see the attachment.
Comments for author File: Comments.pdf
Author Response
Responses to Reviewer #2:
Comment 1: The writing of the paper needs improvement.
Answer 1: Thank you for your kind comments，we have performed speech correction.
Comment 2: Literature review is weak. In-depth analysis of research published recently that is in line with the proposed work needs to be done, and more recently published papers need to be included in the introduction section.
Answer 2: Thank you very much for your suggestions, we have rearranged the literature review to cite more recently published papers.
Comment 3: Need to add some analysis of the experimental results.
Answer 3: Thank you very much for your suggestions, we have increased the analysis of experimental results in Section 3.2 and Section 3.3.
Comment 4: Avoid waste words like "Studies on", or "Investigations on”, “effects of”, “comparison of”, or “a case of” Use specific terms rather than general
Answer 4: We have re-proofread the full English language and revised the non-standard statement.
Comment 5: More sufficient results and comments should be added, when compared to the more recent algorithms.
Answer 5: Thank you very much for your suggestions, we have added experiments and analysis of results in Section 3.3.
Comment 6: Subheadings related to the research gap and novel contribution that are direct implications of the introduction section need to be included.
Answer 6: Thank you very much for your suggestions, we have added novel contribution in the introduction section.
Comment 7: Proposed methodology is not presented well.
Answer 7: Thank you very much for your suggestions, we have reorganized the methods, more details please see section 2. Moreover, we have used an algorithm to describe the method in section 2.1, line 111-119.
Comment 8: The references should be expanded, and more references related within the scope of the article should be added.
Answer 8: Thank you very much for your suggestions, we have searched for relevant references and reorganized them, as well as citing.
- Padilla R, Passos W L, Dias T L B, et al. A comparative analysis of object detection metrics with a companion open-source toolkit [J]. Electronics, 2021, 10(3): 279.
- Xin M, Wang Y. Research on image classification model based on deep convolution neural network [J]. EURASIP Journal on Image and Video Processing, 2019, 2019: 1-11.
- Huang Z, Yang S, Zhou M C, et al. Making accurate object detection at the edge: review and new approach [J]. Artificial Intelligence Review, 2022, 55(3): 2245-2274.
- Kvietkauskas T, Stefanovič P. Influence of Training Parameters on Real-Time Similar Object Detection Using YOLOv5s [J]. Applied Sciences, 2023, 13(6): 3761.
- Barbosa W O, Vieira A W. On the improvement of multiple circles detection from images using hough transform [J]. TEMA (São Carlos), 2019, 20: 331-342.
- Comaniciu D, Meer P. Meer, P.: Mean shift: A Robust Approach Toward Feature Space Analysis [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002, 24(5):603-619.
Thank you for all your valuable suggestions and advice, we are very grateful and will try our best to modify the paper until it meets all your requirements and expectations. Please let me know if any further revision is necessary.
Author Response File: Author Response.pdf
Reviewer 3 Report
Some suggestions/corrections are included in the pdf file
Comments for author File: Comments.pdf
Author Response
Thank you for all your valuable suggestions and advice, we are very grateful and will try our best to modify the paper until it meets all your requirements and expectations. Please let me know if any further revision is necessary. The specific modification instructions have been uploaded as an attachment. Please check it
Author Response File: Author Response.pdf
Reviewer 4 Report
The authors solve the problems of detecting objects in real-time images by modifying the Circle Fitting Algorithm, which showed more stable results compared to the least squares method. However, there are some shortcomings in the work that should be corrected, namely:
1) Line 14: paper -> article! And further in the text
2) Please expand the introductory part, because it is quite concise now. It can be recommended to consider general works on the problem of object detection (10.3390/app13063761, 10.18287/2412-6179-CO-922, 10.3390/electronics10030279).
3) Line 111 - 119: should be presented as a description of the algorithm.
4) Line 200: you need to enter the concept for the abbreviation PLC
5) Line 215: 4G -> 4 Gb
6) For table 1, in my opinion, it is also important to give the sizes of different images.
7) What is plotted along the axes in Figure 5? If this is an error in estimating the diameter, wouldn't it be better to calculate the square of the error?
Author Response
Responses to Reviewer #4:
Comment 1:Line 14: paper -> article! And further in the text
Answer 1：Thank you for your kind comments.We have re-proofed the language of the entire text and corrected the irregular descriptions.
Comment 2:Please expand the introductory part, because it is quite concise now. It can be recommended to consider general works on the problem of object detection( 10.3390/app13063761、10.18287/2412-6179-CO-922 .10.3390lelectronics10030279)。
Answer 2：Thank you very much for your suggestions and recommended papers. We have reorganized the introduction section to cite more recently published references.
- Padilla R, Passos W L, Dias T L B, et al. A comparative analysis of object detection metrics with a companion open-source toolkit [J]. Electronics, 2021, 10(3): 279.
- Андриянов Н А, Дементьев В Е, Ташлинский А Г. Обнаружение объектов на изображении: от критериев Байеса и Неймана–Пирсона к детекторам на базе нейронных сетей EfficientDet[J]. Компьютерная оптика, 2022, 46(1): 139-159.
- Kvietkauskas T, Stefanovič P. Influence of Training Parameters on Real-Time Similar Object Detection Using YOLOv5s [J]. Applied Sciences, 2023, 13(6): 3761.
- Xin M, Wang Y. Research on image classification model based on deep convolution neural network [J]. EURASIP Journal on Image and Video Processing, 2019, 2019: 1-11.
- Huang Z, Yang S, Zhou M C, et al. Making accurate object detection at the edge: review and new approach [J]. Artificial Intelligence Review, 2022, 55(3): 2245-2274.
- Comaniciu D , Meer P . Meer, P.: Mean shift: A Robust Approach Toward Feature Space Analysis [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002, 24(5):603-619.
Comment 3:Line 111 - 119: should be presented as a description of the algorithm.
Answer 3：We have modified the description of the algorithm. See lines 111-119 of the revised version.
Algorithm
Input：T, Tb
（T is the initial position of the circle fit, and Tb is the interval between the points）
Output：Ta
- Ta=T；
- Ta=Ta+Tb；
- If the number of Ta < Maximum number of points
- Tb++;
- Ta=Ta+Tb;
- If the number of Ta < 3;
- T++;
- Ta=T;
- Else Ta performs least squares
Comment 4:Line 200: you need to enter the concept for the abbreviation PLC
Answer 4： PLC is an abbreviation for Programmable Logic Controller and has been supplemented in revised version.
Comment 5:Line 215: 4G -> 4 Gb
Answer 5：It has been modified to 4GB RAM.
Comment 6:For table 1, in my opinion, it is also important to give the sizes of different images.
Answer 6：In Table 1, the operation time of 5 images with different numbers of circular holes in Fig. 2 is presented, with pixels of 1514 each × 1787，1790 × 2185，1737 × 3018，2418 × 3037，2328 × 3037. The above images were all taken from the original image of the punched nickel plated steel strip, ensuring that the parameters of the circular holes are basically consistent. In Fig.2, for the sake of layout effect, different sizes of graphics are scaled, making the size of circular holes appear different.
Thank you very much for your suggestion. If further clarification on this issue is needed, we will try our best to modify the paper until it meets all your requirements.
Comment 7:What is plotted along the axes in Figure 5? If this is an error in estimating the diameter, wouldn't it be better to calculate the square of the error??
Answer 7：Thank you very much for your suggestions. The error standard given by the enterprise is the diameter error within 10um, according to your suggestion, we defined the absolute diameter error as follows
Error= |Standard value -Calculated value| (13)
The hole diameter error curve of punched steel strip of No. 145 is shown in Fig. 5. The abscissa of Fig. 5 represents the number of circles inspected by nickel-plated steel strip of Model 145 punched steel, and the ordinate is the diameter error calculated by three different methods.
The absolute error is used in this article because the error standard given by the enterprise is that the diameter error is within 10um, so the calculation error is directly defined in this article, without using the square of the error.
Thank you for all your valuable suggestions and advice, we are very grateful and will try our best to modify the paper until it meets all your requirements and expectations. Please let me know if any further revision is necessary.
Author Response File: Author Response.pdf
Round 2
Reviewer 4 Report
The authors corrected almost all comments.
However, an English-language reference to source No. 7 in the list of references should be provided.
Author Response
Comments and Suggestions for Authors
The authors corrected almost all comments.
However, an English-language reference to source No. 7 in the list of references should be provided.
Answer：Thank you for your kind comments. An English-language reference to source No. 7 in the list of references has been provided.
- Andriyanov NA, Dementiev VE, Tashlinskii AG. Detection of objects in the images: from likelihood relationships towards scalable and efficient neural networks[J]. Computer Optics.2022,46(1): 139-159.
感谢您的所有宝贵建议和建议。
Author Response File: Author Response.pdf
