Surface Heat Monitoring with High-Resolution UAV Thermal Imaging: Assessing Accuracy and Applications in Urban Environments
Round 1
Reviewer 1 Report
Comments and Suggestions for Authors
The abstract of the article is enough.
Introduction should be improved. Literature review is not enough.
The language of the article is not academic. English proofreading is necessary!
The methodology of the article is not clearly stated and this makes the article very weak.
The methods and parameters used in the article are not supported by the literature. For what purpose did you do the different combinations? How do they compare with studies in the literature?
The limitations of the paper are not sufficiently explained. You did not mention the effect of varying temperature values of the day on the studies. Please direct me if I have not seen it.
The resolution of all figures is not enough. Please increase the dpi.
Line 78: "(Error! Reference source not found.)." ?????
Line 84: In Line 78 and Line 85 you say you have 3 targets but here it says 4 targets.
Line 189-191: It is not clear how you convert pixel values to temperature values!
Author Response
Please see the attachment. Thank you
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsComments for author File: Comments.pdf
Author Response
We have deleted the "4)" which was left over from an earlier draft. The reference not found error is not in the manuscript from our side--perhaps this is an uploading error.
We thank the reviewer for their time and feedback.
Reviewer 3 Report
Comments and Suggestions for AuthorsThe study effectively establishes three primary experimental objectives in an urban setting:
1. Evaluate the accuracy of the FLIR Vue Pro R camera mounted on an unmanned aerial vehicle (UAV).
2. Demonstrate the practical application of UAV technology in an urban environment, with a particular focus on determining the optimal flight altitude for accurate temperature measurements.
3. Compare the surface temperatures of various urban materials under shaded and unshaded conditions."
This reviewer acknowledges the difficulties implied in this type of experimental study and the merit on the part from the participants that has been involved. The reader also appreciates the clear and thorough presentation of measurements and the appropriate use of statistical analyses.
Despite its strengths, the study could be further enhanced by addressing certain areas for improvement:
- In-depth specifications, including the spectral range, spectral response, temperature accuracy, and resolution, should be provided for both the FLIR Vue Pro R 640 thermal camera and handheld FLIR E6-XT infrared imaging camera. This information is crucial for a more comprehensive understanding of the measurements and their interpretation.
- The study's lack of consideration for emissivity and atmospheric correction is a significant oversight. These factors significantly impact the accuracy of thermal imaging measurements, and their omission raises concerns about the reliability of the study's findings. It is crucial to account for emissivity, which is a measure of how effectively a surface radiates energy. Only for a blackbody, presumably your black tile, brightness temperature is equal to the actual temperature. For non-blackbody surfaces, such as the gray and white tiles used in the study, the measured brightness temperature is not equal to the actual temperature. Additionally, atmospheric correction is essential at higher altitudes, particularly above 30 meters regarding literature, as the amount of atmospheric interference increases with distance. Without proper emissivity and atmospheric corrections, the study's temperature measurements may be significantly inaccurate. These issues should be addressed in the study, at least from a theoretical point of view if no application is posible, as one of the most important challenges in thermal imaging is the appropriate calibration and atmospheric/emissivity correction of the cameras that may be used at field.
- The practice of equating brightness temperatures with actual temperatures is a significant methodological flaw that must be addressed. In particular, the mixing of contactless grass temperature readings from the FLIR handheld with contact temperatures from the thermal tiles to derive a model for converting image pixels into observed temperatures is highly problematic. This approach is fundamentally flawed because it combines measurements from different sources without accounting for the emissivity of the surfaces being measured. The resulting model, presented in lines 143-144 and used extensively throughout the paper, lacks credibility due to its inherent inconsistencies. Therefore, its application in the study is highly questionable and should be reconsidered.
- Another critical issue that requires further clarification is the specific radiometric calibration procedure employed by the FLIR Vue Pro R camera, particularly whether it involves an internal temperature calibration process and a stabilization period after switching on. If so, the authors must provide detailed information about the camera's configuration and how this calibration procedure was implemented. Many cameras utilize unique and proprietary onboard calibration systems to reduce image noise and optimize image quality for each scene captured. Radiometric calibration is particularly sensitive to environmental factors such as air temperature, humidity, and the internal temperature of camera components. Additionally, the study's discussion of image formats is incomplete. It is not clearly explained whether users have the option to adjust environmental conditions, target-related parameters, or whether pre-processing is handled automatically by internal factory processes. A thorough explanation of the differences between rjpg and tiff image pixel factory conversion models is also lacking.
- The handheld FLIR E6-XT infrared imaging camera appears to be the most accurate radiometer, as its readings closely match the contact temperatures of the tiles. However, it's crucial to acknowledge that contact and contactless temperature measurements may differ due to emissivity differences. The black tile provides the most reliable comparison point, resembling a blackbody more closely than the gray or white tiles. It would be beneficial to provide the tile materials and contact thermal sensor specifications.
- The study findings suggest that the FLIR Vue Pro R camera may be less accurate than the handheld radiometer, as the brightness temperatures measured by the two devices exhibit greater discrepancies than those for the hanheld and tile comparison. These differences can be attributed to various factors, including emissivity differences, atmospheric interference, and internal radiometric calibration settings. To establish the most realistic accuracy of the UAV instrument, a comprehensive analysis of all these factors is warranted.
Author Response
Please see the attachment. Thank you
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsThe corrections I suggested have been taken into account. The article can be published as is.
Author Response
We thank the reviewer for all of their time and feedback.
Reviewer 3 Report
Comments and Suggestions for AuthorsThank you for the continued effort you have put into improving the study. The manuscript is nearing publication readiness, with only a few minor issues remaining that I encourage the authors to address. These are outlined in the following comments to ensure the final manuscript is as strong as possible.
Line 115: Please, provide all measurements in international units. Please check the thickness value provided because it sounds very thick. Please, check for possible errors.
Table 1 and 2: Magnitudes are not needed to be expressed in Imperial Units. Consider the international circulation of the Remote Sensing Journal.
Line 188: Not sure what you understand by kinetic temperature. I would use the term "skin temperature" for the actual temperature of the uppermost layer of the surface, measured in degrees Celsius (°C). Besides, I would use the term "radiative temperature" to express the apparent temperature derived from the amount of infrared radiation emitted by the surface, normally measured in degrees Kelvin (K) since when converted to radiation and this radiation is divided by the emissivity, it will be related to the skin temperature.
Equation 2: It needs a reference, since [30] seems to apply only for the emissivity concept.
Line 192: It is difficult for the reader to understand the procedure to estimate emissivity using black electrical tape. It would help if you elaborated on this concept a little more. Also, you should indicate that your approximation did not consider any interfering radiation reflected from the celestial vault. This is not atmospheric correction, it is a correction that comes from the reflected radiation when surfaces do not behave as black bodies.
Author Response
Please see the attachment. We thank the reviewer for all of the time and feedback.
Author Response File: Author Response.pdf
