Vegetation Classification in Urban Areas by Combining UAV-Based NDVI and Thermal Infrared Image
, Gyeonggyu Kim 2, Gyeongjo Min 2, Minju Kim 3, Seunghyun Jung 3, Jeewook Hwang 4,* and Sangho Cho 2,5,*Round 1
Reviewer 1 Report
This manuscript used NDVI from multispectral image combined with thermal image to extract vegetation areas in urban areas. Thermal information is used to distinguish the vegetation areas from the urethane coated areas which cannot be distinguished by NDVI along. The classification method is very simple, which is only based on thresholds. the innovation is low, and the structure also needs to be improved.
1. The spectral information of Canon S110 NIR camera needs to be provided in section 2.3.
2. The data acquired by ASD (not ADS) should introduced in Section 2.
3. Figure 6 can be deleted since it provided little information.
4. the structure of the results part is suggested to be reconstructed.
For example, 3. Results and discussion
3.1 the spectral signatures of different types of covers
3.2 Classification results based on NDVI
3.1 Classification results based on NDVI and thermal infrared image
5. In Line 217-218, There is no quantity accuracies of the vegetation classification results based on different NDVI thresholds, which is not convincing.
6. in Line 237-253, this part is suggested to move to method part. What is the purpose of this part? Does it provide evidence that the spectral reflectance of the vegetation and the urethane coated areas are different? but from Figure 13, it can be spotted that there have obvious spectral differences between them. especially between 700-900nm. through they may have similar NDVI which can not be directly figured out from these spectral signatures.
7. Line 257-258, “As a result of the analysis, NDVI 257 values of basketball courts and waterproof coating roofs were similar to those of grass 258 with slightly lower vegetation.” You need to first specify the bands you use to calculate NDVI, and then list some statistic results of the NDVI of different covers to support you point.
8. Line 283-286: The classification method is really simple which is only based on thresholds. I recommend authors could try to use more complicated classification methods based on less sensors (e.g. multispectral image solely or thermal image solely).
9. Table 8: the unit is missing.
Author Response
Dear Reviewer 1,
We appreciate your interesting and informative review. Your insightful review helped us make our manuscipt more comprehensible and logical.
Please see the attachment.
We appreciate the review and your academic contribution.
Best regards,
Authors
Author Response File: 
Author Response.pdf
Reviewer 2 Report
This study proposes to develop a technology to effectively extract vegetation information from UAV based images. Its main contribution consists in combining UAV images captured with multispectral and thermal infrared cameras for urban areas including basketball courts and the waterproof coating roofs covered with urethane.
The document is easy to read and follow.
The English needs minor spell checking.
The document is well supported with references.
The subject of the paper has great potential of application.
The proposed work main weakness is the lack of information about the behavior of the methodology in various weather conditions.
The sentence in line 61 and 62 is not very clear. Please revise.
In Figure 1 did you mean “Compares the classified vegetation area…”?! Please correct. Also did you mean “…misclassification area such…”. Please correct.
In Figure 2 it is not clear why the authors included in the “Legend” the Red, Green and Blue. It seems that the only relevant information in the “Legend” is the “Study Area” defined in red contour in the image. Please correct this issue in all the figures.
Please correct in Table 1 the “Longitudinal overlap” of “ThermoMAP”.
Authors should correct the caption of Table 4 “Vegatation”.
Below Table 4 authors described the time of year (July) when the image acquisition for the experiments was conducted. Authors should also include a description about the weather conditions when the image acquisition took place.
Author Response
Dear Reviewer 2,
We appreciate your interesting and informative review. Your insightful review helped us make our manuscipt more comprehensible and logical.
Please see the attachment.
We appreciate the review and your academic contribution.
Best regards,
Authors
Author Response File: Author Response.pdf
