Next Article in Journal
Elevating Academic Advising: Natural Language Processing of Student Reviews
Previous Article in Journal
Design and Implementation of Nursing-Secure-Care System with mmWave Radar by YOLO-v4 Computing Methods
 
 
Review
Peer-Review Record

Deep Learning for Automated Visual Inspection in Manufacturing and Maintenance: A Survey of Open- Access Papers

Appl. Syst. Innov. 2024, 7(1), 11; https://doi.org/10.3390/asi7010011
by Nils Hütten *, Miguel Alves Gomes, Florian Hölken, Karlo Andricevic, Richard Meyes and Tobias Meisen
Reviewer 1: Anonymous
Reviewer 2: Anonymous
Appl. Syst. Innov. 2024, 7(1), 11; https://doi.org/10.3390/asi7010011
Submission received: 15 November 2023 / Revised: 27 December 2023 / Accepted: 16 January 2024 / Published: 22 January 2024

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This survey paper demonstrates commendable work in synthesizing a wide range of studies through a systematic approach. However, there are areas where improvements could significantly enhance its value:

 

- In Table 4, there is an issue with overlapping text, which hinders readability and understanding. 

 

- The paper predominantly focuses on high-level, quantitative overviews in a top-down analytical framework. While this offers valuable insights, the paper could benefit from augmentation of in-detail analysis, particularly in the context of Deep Learning (DL) applications in manufacturing and maintenance. 

 

Specific suggestions include:

Detailed DL Analysis: An in-depth exploration of the characteristics of DL methods used in various application cases of manufacturing and maintenance would be beneficial. This should encompass the nature of the DL algorithms adopted, their unique attributes, and their suitability for use-case scenarios.

 

Target Data Set Examination: A closer examination of the data sets used in these applications, including their specific features and challenges, would add depth to the analysis.

 

Complex Analysis Approach: There's a need for a more comprehensive analysis that intertwines application characteristics, dataset features, technical methodologies, and evaluation metrics. This complex analysis should recognize that the significance and semantic of metric values can vary depending on dataset and application characteristics. Rather than relying solely on statistical processing of numerical values, a multifaceted approach would provide a richer, more insightful understanding of DL applications in these fields.

 

By addressing these areas, the paper could offer a more balanced and in-depth perspective, thereby enhancing its contribution to the field.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

In this paper, the authors present a survey of DL methods for visual inspection tasks.

There are several remarks on the paper.

1. The main remark on the survey is "we only considered open-access publications" this is a big drawback. The paper should then have a title "...: A Survey of Open Access Papers"

2. There is only one reference for the tiling industry. Missing some papers in tiling industry use case that must be mentioned like from the authors Karimi MH, Asemani D. Surface defect detection in tiling industries using digital image processing methods: Analysis and evaluation. or Bruno Zorić , Tomislav Matić, Željko Hocenski Classification of biscuit tiles for defect detection using Fourier transform features.

3. Papers for pest classification eg. Čirjak, D., Aleksi, I., Miklečić, I., Antolković, A. M., Vrtodušić, R., Viduka, A., Lemic, D., Lemić, D., Pajač Živković, I. (2023). Monitoring System for Leucoptera malifoliella (O. Costa, 1836) and Its Damage Based on Artificial Neural Networks. Agriculture 13(1): 67.

4. Table 4. has an overtype in the 5th row.

5. Additional analysis should be given about imbalanced data sets because this is relatively common in the industry

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The revised submission has been improved and is considered to be published as it is.

Reviewer 2 Report

Comments and Suggestions for Authors

If the authors do not wont to include papers that are not open access then they must change the title of the paper and change the abstract so the reader at the beginning knows what types of paper are included in the Survey. As I mentioned in the first review the title must be "... survey of open access papers".

Author Response

Dear Reviewer,

we have updated the manuscript according to your suggestions and changed the title as well as the abstract to reflect that we only considered open access literature.

Best regards!

Round 3

Reviewer 2 Report

Comments and Suggestions for Authors

The authors corrected the paper according to my suggestions.

Back to TopTop