Combining Spectral, Spatial-Contextual, and Structural Information in Multispectral UAV Data for Spruce Crown Delineation
Round 1
Reviewer 1 Report
Please see attachment.
Comments for author File: Comments.pdf
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 2 Report
It is a very good paper with both clarity and necessary details. This paper proposes an individual tree crown delineation method based on the utilization of the spectral, contextual, and structural information from UAV data through a combination of multiple kinds of methods. The methodology is well described and the results show the advantages of their methods over two normally used methods. There are a few suggestions for the authors to be considered:
- For readers to quickly get the big picture of the methodology and to better understand the usage of each method in the whole project, I suggest adding the major method (or methods) in each step in Fig. 2.
- Though this paper emphasizes their method is based on spectral, spectral-contextual, and structural information, I do not really see how the contextual information is computed or used in this paper.
- What is the meaning of Line 255-256?
- There are no black lines in Fig. 6a.
- The green lines in Fig.7 are hard to be seen.
- It seems the ridges better fit the reference crown boundaries than the detected crown boundaries from Fig. 7, why bother to obtain crown boundaries in this manner?
- Line 375: IoU should be no units.
- Black and white lines are not so clear in Fig.8. It needs to use other colors for a better visual view.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf