An Efficient and Accurate Surface Defect Detection Method for Wood Based on Improved YOLOv8
, Fulong Liang 1,2, Bo Wang 2,3,*, Guanghao Zhang 1,2, Yesheng Chen 1,2 and Xiangwei Mou 1,2,*
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsTaking into account the possibility that this manuscript does get published, I think it is appropriate to include some brief comments.
- The manuscript seems somewhat with grammatical/syntax and typographical problems. I leave it to the authors to resolve these copyediting problems by actually thoroughly reading the manuscript. Problems of this sort should definitely not appear in print.
Congratulations on your ongoing/concluded research.
Comments for author File: 
Comments.pdf
Taking into account the possibility that this manuscript does get published, I think it is appropriate to include some brief comments.
- The manuscript seems somewhat with grammatical/syntax and typographical problems. I leave it to the authors to resolve these copyediting problems by actually thoroughly reading the manuscript. Problems of this sort should definitely not appear in print.
Congratulations on your ongoing/concluded research.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for Authors
Review report:
• A brief summary (one short paragraph) outlining the aim of the paper, its main contributions and strengths
Paper presents a novel for detecting wood surface defects using a modified YOLOv8 model. Authors presents as main contributions the integration of the Dynamic Weighting Receptive Field (DWR) module and Deformable Large Kernel Attention (DLKA) into the C2f module of YOLOv8, enhancing multi-scale feature extraction and fusion. The detection head of YOLOv8 is upgraded to a dynamic detection head, and an additional dynamic detection head is added in the shallower P2 layer to improve small target detection. Additionally, the original CIOU loss function is replaced with the MPDIOU loss function to facilitate faster and more efficient regression. Dataset used in the paper is derived from VSB-Technical University of Ostrava and is enhanced with data augmentation techniques to improve model robustness and prevent overfitting. In experimental results, authors show a significant performance improvement, with a 5.5% increase in mean Average Precision (mAP) over the baseline YOLOv8n model. Authors mention specific improvements in detecting Live_Knot, Resin, Dead_Knot, Knot_missing Crack classes defects, second best for Morrow class and forth place for Knot_with_crack class. The proposed enhanced model provides a robust, efficient, and accurate solution for wood surface defect detection, suitable for practical applications in the wood processing industry.
• General concept comments
Main observations are:
- bibliography contains 39 references, 2 older than 10 years, 12 references from last 10 years and 25 from last 5 years
- all parameters used in proposed algorithm should be explained
- results in tables should be emphasized
- English language could be improved
• Specific comments referring to line numbers, tables or figures that point out inaccuracies within the text or sentences that are unclear.
- line 422 - in Table 2. Experimental Environment Configuration - "Random Access Storage" should be changed to Random-access memory
- line 436 - in Table 3. Ablation Experiment Results for YOLOv8 Variants - best results on each row should be emphasized
- line 478 - in Table 4. Comparison results with other algorithms - best results for mAP% and AP% for each class should be emphasized
- line 414 in Algorithm1. Inner-IoU and MPDIoU as bounding box losses - all parameters from steps 1 - 11 should be explained
• Is the manuscript clear, relevant for the field and presented in a well-structured manner?
Manuscript is relevant for the field and well presented.
• Are the cited references mostly recent publications (within the last 5 years) and relevant? Does it include an excessive number of self-citations?
Cited references are new, out of 39 references, 2 references are older than 10 years, 12 references are from last 10 years and 25 references are from last 5 years
• Is the manuscript scientifically sound and is the experimental design appropriate to test the hypothesis?
The experiment design is appropriate to test the hypothesis.
• Are the manuscript’s results reproducible based on the details given in the methods section?
More information should be added in order for the results to be reproducible.
Comments on the Quality of English Language
English language can be improved.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
Comments and Suggestions for AuthorsThank you for your effort, in the revised version observations made in the first round review were addressed and recommendations have been implemented.
Comments on the Quality of English LanguageEnglish language was improved in the new version.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
