Research on Algorithm for Authenticating the Authenticity of Calligraphy Works Based on Improved EfficientNet Network
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThe paper presents research on automatic recognition of forgery in calligraphy works.
I have a few concerns regarding the paper.
The research presented in the paper can be divided into two steps. In the first step bounding boxes of characters are detected and in the second step deep learning is used.
Regarding the first step I have a major concern such that authors did not compare their method of creating bounding boxes to other existing method. The topic is not new therefore a lot of other method were proposed.
Apart from that it is not clear for me how data augmentation (Sect. 2.3.) was applied to images. Was is performed manually or with the use of some automation?
In Sect. 2.2.1 and 2.2.2. Authors should provide references to papers presenting equations included in these sections.
Regarding the second step in which neural network were used it is not clear how the network proposed by authors of the paper differs from existing networks.
Moreover information about the number of images used for testing with regard to number of test images is not presented in the paper.
Authors should also provide references to papers describing different algorithm which they included in tables 1, 2 and 3.
Author Response
Dear reviewer, thank you for your careful review, we have made a point-to-point reply to your suggestions, please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for Authors1.Describe how data addition happens in your algorithm?
2.If the symbols are handwritten, then there may be deviations in the angle of inclination of the main strokes, the thickness of the strokes - how is this taken into account?
3.Your literature is mostly three years old. I think it should be eliminated, because there is a lot of development in the field of artificial intelligence
4. Calligraphic fonts are very different. They are distinguished by new forms and drawings. In this study, it is not clear how this or that letter is identified.
Author Response
Dear reviewer, thank you for your careful review, we have made a point-to-point reply to your suggestions, please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsI have no further comments to the paper.
Reviewer 2 Report
Comments and Suggestions for Authorsno comments
