Seedling-YOLO: High-Efficiency Target Detection Algorithm for Field Broccoli Seedling Transplanting Quality Based on YOLOv7-Tiny

Round 1

Reviewer 1 Report

Dear

Editor I attach my review report, suggestions and recommendations go in the attached document as comments.

Regards

Comments for author File: Comments.pdf

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

Robotization of machine work in agriculture is in line with the implementation of the concept of Agriculture 5.0. Thus, the topics discussed are very current and important for practice. The authors used to control the planting of seedlings of broccoli (it is appropriate to add the Latin name) one of the best-known algorithms for YOLO object identification in the latest version 7.

This tool used for computerized object detection is well-known and successfully used. It is used in robotics, equipment monitoring, self-driving cars and many other areas (many works are included in the MDPI Publishing House). It allows us to find and recognize certain things in a photo or video with high speed and accuracy.


Can this method also be applied to other field work in broccoli cultivation?

On what basis was such a low number of 29.7 frames per second (FPS), which older versions of Yolo already meet? YOLOv5, for example, can process images at up to 1,000 frames per second on a single GPU.

What happens when there are identified planting deficiencies? If Nothing, then why does anyone need to know?

no concrete

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comment

1. Adapt the abstract and introduction.

2. summarize the findings more explicitly and broader the applicability of the methods.

3. In a few places, full stops are missing. For eg. the first para in the introduction.

4.  Bit expand the literature including the work on drone for precision agriculture [1]

[1] https://doi.org/10.3390/rs15092450 

5. better to provide some analysis to show parameter vs performance among the models M1-M8

6. Shed some light on attention-based lightweight models in precision agriculture [1] 

[1] https://doi.org/10.1371/journal.pone.0264586

7.  Since this work is aiming for a lightweight model for edge devices, the author may consider YOLO Nano [1] to experiment.

[1] https://doi.org/10.48550/arXiv.1910.01271 

8. Most importantly to replicate the research and use it for other researchers, authors may make available the source code of their models on GitHub or similar.

In a few places, full stops are missing. For eg. the first para in the introduction.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Dear

Editor The New version of mansucript its ok

Regards

Reviewer 3 Report

The author addressed all of the comments. Thank you for putting in the effort to get the paper better.