Next Article in Journal
On Neighborhood Structures and Repair Techniques for Blocking Job Shop Scheduling Problems
Previous Article in Journal
Tensor-Based Algorithms for Image Classification
 
 
Search for Articles:
Title / Keyword
Author / Affiliation / Email
Journal
Article Type
 
 
Advanced Search
 
Section
Special Issue
Volume
Issue
Number
Page
 
Journals
Algorithms
Volume 12
Issue 11
10.3390/a12110241
Review Report
algorithms-logo
Submit to this Journal Review for this Journal Propose a Special Issue
► Article Menu

Article Menu

share Share announcement Help format_quote Cite question_answer Discuss in SciProfiles
Article
Peer-Review Record

Fingerprints Classification through Image Analysis and Machine Learning Method

Algorithms 2019, 12(11), 241; https://doi.org/10.3390/a12110241
by Huong Thu Nguyen 1,2,* and Long The Nguyen 3
Algorithms 2019, 12(11), 241; https://doi.org/10.3390/a12110241
Submission received: 3 October 2019 / Revised: 7 November 2019 / Accepted: 8 November 2019 / Published: 11 November 2019
(This article belongs to the Special Issue Algorithms for Content Based Image Retrieval)

Round 1

Reviewer 1 Report

First of all I would like to congratulate the authors because I think it is a good machine learning and image processing application. So, as far as the preprocessing algorithm is concerned, I think it is correctly described. Anyway, I think there are several aspects that should be incorporated to improve the quality of the manuscript and show the power of the algorithm.
1) The figures must be improved in quality.
2) Why SVM and RF? This must be justified and incorporate neural networks in the comparison because it is a very different approach to SVM, as well as their training algorithms.
3) The results, in order to show the power of the image preprocessing that is performed, must be compared with deep learning models incorporating the image directly into the model. If not, justify why not.

 

Author Response

First of all, the authors are grateful to the referee for careful reading of the paper and valuable suggestions and comments. There are our corrections to improve paper. 

Please see the attachment.

Thank you so much!

Best regards, Thu Huong & The Long

Author Response File: Author Response.docx

Reviewer 2 Report

This paper presents the application of machine learning methods to develop fingerprint classification algorithms based on singularity feature.
The paper presentation is not adequate and there is no novelty.
Comments:

1)They must use novel methods such as deep learning techniques which are suitable especially for image processing.
2)Besides the classification accuracy, they must use other performance measures such as F-measure, ROC area and Kappa statistics.
Moreover, they must apply some statistical tests such as t-test.

3)They must give their accuracy results in the abstract as well.

4)There should be a comparison section with the state of the art.

5)Besides RF and SVM they must use other benchmark classifiers such as ANN, deep learning, k-NN and Decision Tree Classifiers.

Author Response

First of all, the authors are grateful to the referee for careful reading of the paper and valuable suggestions and comments. There are our corrections to improve paper. 

Please see the attachment.

Thank you so much!

Best regards, Thu Huong & The Long

Author Response File: Author Response.docx

Back to TopTop