Automatic Recognition of Microstructures of Air-Plasma-Sprayed Thermal Barrier Coatings Using a Deep Convolutional Neural Network

Round 1

Reviewer 1 Report

In this article, a machine learning-based approach (DCNN) was established to quantify the microstructure of air plasma spray (APS) TBs based on two-dimensional images. After the training applied in the study, it was observed that DCNN could recognize pixels in SEM images of typical APS TBCs with high and acceptable accuracy. As a result of this study, it was stated that using a small amount of SEM images, APS is sufficient to characterize the microstructure of TBs. It also demonstrated the originality and importance of the article, which may be sufficient to train a DCNN providing a viable method offered by the authors.

As a result of the evaluation, the publication of this study was deemed scientifically appropriate. In general, it is sufficient to review the English written language again. In addition, it would be more beneficial to increase the number of citations in the introduction of the article a little more.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

The authors report on the course of research aimed at creating a tool for determining the properties of thin coatings. They show a method based on machine learning, where the input data are images from an electron microscope. The manuscript is interesting, well organized and printable. However, there are a few places to check before printing. The only thing that surprises me a bit is that the accuracy is given as a percentage with one decimal place. For example, on line 216, 217, 218 and others. I understand that accuracy is the result of estimation, not precise calculation. On the labels next to the bars in Figure 9 and others OK of course. It is a bit surprising that the authors thank the sponsor in line 388, since they mentioned it in line 382. In "Acknowledgments" we direct our gratitude for substantive, organizational or technical help.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

In this paper, the authors have proposed a deep convolution neural network model to perform pixel-wise automatic microstructure recognition of SEM images of air plasma spray thermal barrier coatings. The paper has a merit in its originality and novelty of the proposed model, and the results are well presented with concise analyses and conclusions. I recommend this paper for publication in Coatings journal. 

Author Response

Please see the attachment.

Author Response File: Author Response.docx