Infrared Image Small-Target Detection Based on Improved FCOS and Spatio-Temporal Features
Round 1
Reviewer 1 Report
See the attached.
Comments for author File: Comments.pdf
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 2 Report
The authors proposed the approach based on the improved FCOS and spatial-temporal features to detect the infrared image small target. Several comments are given as follows:
- Detecting the small objects in the infrared images is a difficult job but the authors provide an efficient way to detect the small objects. However, the small objects in the infrared images are supposed to define those small objects that are moving. For those small objects without moving, the proposed approach could fail.
- The preprocessing step is introduced maximum filter to enhance the small objects. Can the authors provide the explanations about the reasons to choose the largest neighbors instead of selecting the center pixel in 3 by 3 window? In doing so, it not only can depress the spot noise but also smooth the small objects.
- The authors did a great work to compare the performances of different approaches but did not mention the processing speeds of employing different algorithms. The authors are supposed to make a comparsion.
It is not recommended to publish the paper at this moment.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
The authors addressed my concerns. The manuscript may be published now.
Author Response
Thanks to the reviewers and editors for their help, we have carefully revised according to the comments, hoping to meet the publication requirements. Modifications to the conclusion have been highlighted.
Please see the attachment for revised paper
Author Response File: Author Response.pdf
Reviewer 2 Report
The authors made a lot of modifications. No more comments are given.
Author Response
Thanks to the reviewers and editors for their help, we have carefully revised according to the comments, hoping to meet the publication requirements. Modifications to the conclusion have been highlighted.
Please see the attachment for revised paper
Author Response File: Author Response.pdf