Particle Swarm Optimisation in Practice: Multiple Applications in a Digital Microscope System

Round 1

Reviewer 1 Report

The paper describes four applications of the conventional PSO algorithm in the domain of electron microscopy. The paper is well written. Nevertheless, I miss the main contribution of the paper as it uses a well-established single-objective optimizer to solve relatively simple problems (having a low number of decision space variables < 10).

The paper seems to be too long taking into account the presented novelty. In my opinion, the example "Colour Reproduction Accuracy" (sec. 5) could be omitted from the manuscript. It is the longest example and the use of PSO does not bring any benefits compared to the Moore-Penrose pseudo-inverse method, which is also more stable.  

Nevertheless, the paper might be interesting to some readers of the Applied Sciences journal as it highlights the beauty of global optimization algorithms. They can be used with low efforts of the user to solve a large set of relatively complex tasks. 

Please, see the minor remarks and found typos in the enclosed document. 

Comments for author File: Comments.pdf

Author Response

We thank the reviewer for his careful reading and the many good suggestions. This is a very helpful review. In detail, we changed these issues:

The paper seems to be too long taking into account the presented novelty. In my opinion, the example "Colour Reproduction Accuracy" (sec. 5) could be omitted from the manuscript. It is the longest example and the use of PSO does not bring any benefits compared to the Moore-Penrose pseudo-inverse method, which is also more stable. 

We have shortened the paper by leaving out or reducing images. With respect to sec. 5, we would like to point out that PSO brings benefits if outliers are allowed. We state in sec. 5:

"Pseudo Inverse Matrix produced a consistent error of 310. The PSO algorithm produced errors from 308 to 310. Allowing a single outlier in the PSO algorithm reduced the error to 274, an 11.3\% decrease in the colour reproduction error."

and

"Three areas are highlighted by yellow border where the colours have noticeably improved with the PSO Algorithm. The PSO algorithm found a colour correction that gave an equal or better result than the Pseudo Inverse Matrix 100 times out of 100 times it was run."

Nevertheless, the paper might be interesting to some readers of the Applied Sciences journal as it highlights the beauty of global optimization algorithms. They can be used with low efforts of the user to solve a large set of relatively complex tasks.

This is our intention with the paper. We do not claim groundbreaking new methods or similar.

Please, see the minor remarks and found typos in the enclosed document.

We have corrected all those remarks. We do not comment in detail on the minor changes. More significant issues follow below:

We extended the abstract to have more results in it.

In eq. 1/2 the j is not a temporal index. We have added "new" to variables to show the update.

We have specifie the type of noise in 3.1.1

We added results for the point cloud experiment.

For benefits of PSO in section 5, see above.

The symbol in eq. 14 is a divions symbol, we replaced it by /, which is more common.

Reviewer 2 Report

Ryan et al. presented their work entitled "Particle Swarm Optimization in Practice: Multiple applications in a digital microscope system". In this work, particle swarm optimization is applied to solve several problems in digital imaging inspection.  Height measurement, image stitching, color reproduction, and distortion correction are demonstrated.  Industrial sample inspection is a critical component of quality control and process optimization.  Thus, new solutions and approaches to improve the inspection process is valuable.  This work demonstrates a good example of the implementation of particle swarm optimization. Overall the manuscript is well-written and the presented data supports the claims of the authors.  I have the following minor comments.

* Abstract should be extended to include the major findings and significance of the results. 

*The presented literature in the introduction section can be increased to include the most recent example of the particle swarm optimization and digital microscope inspection solutions.

*Page 2, line 33: there is a typo "am"  should be "an"

* the text in some of the figures are too small to read such as Figure 14 and 15. 

 

Author Response

We thank the reviewer for his helpful comments. In detail, we changed these issues:

Abstract should be extended to include the major findings and significance of the results.

We have extended the abstract.

The presented literature in the introduction section can be increased to include the most recent example of the particle swarm optimization and digital microscope inspection solutions.

We have added some at the end of sections 2.1 and 2.2

Page 2, line 33: there is a typo "am"  should be "an"

Corrected

The text in some of the figures are too small to read such as Figure 14 and 15.

We have changed those figures to not contain the text, which was not relevant, any more.

Round 2

Reviewer 1 Report

The manuscript was improved. It still contains some minor flaws. Some reference numbers are displayed as "?" symbols. Fig. 13 is not properly displayed.

Please, improve these formal issues before you submit the final version.

Author Response

Thank you for your comments. We have fixed the references (we found two on page 16). Fig. 13 seems to display fine for us. Could we perhaps ask the editorial office to liaise with us about any problems? We have made the figure the full width of the text, perhaps that solves the issue, but otherwise we would need some indication of the problem.