OTSU Multi-Threshold Image Segmentation Based on Improved Particle Swarm Algorithm
Round 1
Reviewer 1 Report
The authors have discussed OTSU multi-threshold image segmentation. The following are my comments.
1. The results must be quantified in the abstract.
2. Why PSO and its improved variant for image segmentation? There are plenty of metaheuristic algorithms available in the literature.
3. Compared to OTSU, fuzzy entropy-based multi-level segmentation is very popular, but why do the authors consider OTSU? Refer to 10.1016/j.bspc.2021.103401.
4. The authors should review the related and recent papers to update the literature study. For instance, 10.1002/er.8011, 10.3390/s22030855, 10.1049/rpg2.12475, 10.1155/2019/2482543, etc.
5. The authors should prove the exploration and exploitation ability of the improved PSO by applying and testing benchmark test functions.
6. Literature study needs improvement in many aspects.
7. The authors should conduct a comprehensive simulation study with multiple images. The segmented image of all algorithms needs to be included in the paper for comparison. In addition, histogram levels need to be presented.
8. List the fitness value achieved by all algorithms. The mean and STD of all performance metrics need to be included.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 2 Report
1. How will your study benefit future researchers?
2. What are the limitations of your study?
3. A section for the assumption is also required to show what authors assume to develop this manuscript.
4. Extensive English corrections are required for better understanding.
5. Novelty and study findings should be clearly presented in the Introduction section
6. Graphical abstract must be included.
7. Some of the references are too old. Although they are famous references in this area, it would be better to cite the state-of-the-art viewpoint in this topic, and the new insights into this solved problem.
8. Literature survey needs improvement. Authors must add latest research articles like a) Optimal placement of TCSC and SVC for reactive power planning using Whale optimization algorithm (b) Fuzzy-Based Shunt VAR Source Placement and Sizing by Oppositional Crow Search Algorithm (c) Application of Sine–Cosine Optimization Algorithm for Minimization of Transmission Loss
9. The English language needs a revising before submitting Revised Version of the paper. The language of the manuscript should be largely polished before the final acceptance by a language editor.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 3 Report
This paper mainly discussed particle swarm algorithm, and the algorithm was improved and applied to OTSU multi-threshold image segmentation. It is a classic topic, and its contribution is trivial.
However, maybe there exists some interesting thing, and if the authors can explain the questions which is following,
1. Eq.2 may be wrong, please check it.
2. The symbols in Eq.5 L, p are not explained in the text.
3. Section 4.2 is not clear. I cannot understand what the author said.
4. The figures in figure 4 are too simple. Would you give some more interesting one?
5. The results of section 5 are just about some figures. It is better to give a engineering example.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
The authors have not improved the literature study. In addition, the authors are suggested to test the algorithm with a greater number of images, but they ignored it. Address all the comments and resubmit the paper.
Author Response
Please see the attachment.In addition, the marked version has also been uploaded and submitted in the compressed package.
Author Response File: Author Response.docx
Reviewer 3 Report
The paper is revised and it seems to be better.
However, in Page 18, all the segmented photos were common, and had no sense. Due to the trivial contribution of the paper, it is necessary to show a meaningful photo or example in the paper. Otherwise, the paper is hard to draw the readers' attention.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Round 3
Reviewer 1 Report
I am satisfied with the responses provided. I hope the paper is improved from its original version.
Reviewer 3 Report
After several revising, the paper seems to be better than the previous version, and I have no words to reject it.