A General Relative Radiometric Correction Method for Vignetting Noise Drift
, Maosheng Chen 1, Dong Wang 1, Jinyan Cao 1 and Xiyan Cai 1
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsIn abstract, CAM is spelled out as "color aberration metrics" whereas in line 138 it is “chromatic aberration metrics” - please be consistent.
What does function “profit” means In line 265, equation 7?
In line 317, what does “Remote sensing images have the same characteristics on the line” imply? This sounds like a very general statement for remote sensing data which can’t be the case. Please rewrite this sentence to avoid ambiguity.
In line 324, please specify the size of the small pieces.
In line 409, figure 19, please correct the caption (should be vegetation instead of water).
In line 413, figure 20, please correct the caption (should be city instead of vegetation).
Author Response
Thank you for your review. Please see the attachment.
Author Response File: 
Author Response.docx
Reviewer 2 Report
Comments and Suggestions for AuthorsThis paper proposes a relative radiometric correction method for the vignetting. It has good reference value for removing the vignetting phenomenon in satellite images in the future. However, there are still some issues with the method introduction and result analysis in this article, and it is recommended to revise and review it again. The specific issues are as follows:
1) Line61 to Line90, the advantages and disadvantages of the method can be introduced together in introducing the method in the introduction.
2) Line 157, the flowchart of Figure 1 is unclear. The DDSN is Divided into the stable area and changing area, then how to analyze them separately? Additionally, what is the relationship between imaging time, noise drift, and DDSN?
3) Line 162, incomplete analysis of Figure 2. Can the position differentiation of adjacent sensors be seen? It is best to distinguish the left sensor from the right sensor using a red box.
4) Line 181 states that 56 images were imaged within 130 seconds, while Figure 3 only shows 4 images, which is inconsistent with the two statements. In addition, there is a lack of detailed introduction to the Jilin-1 GF03D28 satellite, such as how many pixels, how many sensors, and what is the width of each vignetting?
5) Line 193, whether all images in Figure 4a come from 56 images within 130 seconds, is not provided in detail in the text.
6) Line 201, please provide specific instructions on how 50% of the energy is obtained.
7) Line 207, the article mentions that the noise of each pixel is independent, and changes with the imaging time. Please provide specific reasons and supporting materials.
8) Line 214, is Table 1 the parameters of the DDSN or the parameters of satellite? What do Integral Number and Gain mean? Why is the value of Gain 2?
9) Line225 and Line227, what is the difference between Figure 6a and Figure 7a? How to distinguish between valid and invalid data? What is the relationship between Figure 6a and Figure 6b? Why are the pixel positions in Figures 6b and 7b inconsistent? Why is there only 200 pixels in Figure 4, while the pixel numbers in Figures 6 and 7d are 9500 and 7000, respectively? The annotation of Figure 6 and Figure 7 should be modified.
10) Line 235 and Line 237, why Figure 8 and Figure 9 have different pixel numbers. What is the boundary between the changing area and the stable area? How to obtain it? The article states that the grayscale change in the stable zone is less than the 1DN value, but in Figure 8, the grayscale change on the left side of the stable zone exceeds the 1DN value
11) Line242, why is it a linear characteristic? How are scene numbers sorted, how do you choose different scenes, and how do you prove that their grayscale values change linearly?
12) Line246, in Figure 10, there are 6 pixels on the left and 6 pixels on the right, without 10 pixels (in line 239)?
13) Line251, is H in formula 4 the total number of rows for all scenes, or the total number of rows for one scene?
14) Line 265, how to calculate the fitting formula in Formula 7? Can you provide specific explanations or examples? How to obtain the imaging time for formula 7?
15) Line267, please provide a calculation case for pixel gain and intercept first.
16) Line 279, why the pixel number in Figure 11 of is 1600, while Figure 12 is 200? And why there be negative values for gain in Figure 11a? Please provide a reason.
17) Line 268, how can goodness be calculated ?
18) Line307, what does the horizontal and vertical coordinates of the histogram lookup table in Figure 13 mean?
19) Line 316, 3.2.1 should be 2.5.1
20) Line322, RA is related to the uniformity of the image, and the smaller the RA, the more uniform the image is. It is not only related to the relative radiometric correction accuracy of the image. Therefore, RA of different images cannot evaluate the relative radiometric correction accuracy. Therefore, this statement is not accurate.
21) Line 326, where does j in formula 12 come from and what is the value?
22) Line334, is Formula 13 for pre corrected or post corrected images?
23) Line 342, how to convert and obtain LAB color space?
24) Line 374, the annotation of Figure 15 needs to be modified
25) Line386, Figure 16 should indicate the source of the image, the corresponding number of images, and whether it also comes from DDSN images. The annotation needs to be modified
26) Line 397,Figure 17 has 400 pixels, which is different from the previous image such as Figure 12 and Figure 15?
27) Line404-Line420, Figures 18-22 from which satellite of the GF03D series? Which band
28) Line408, Figure 19 should be vegetation,
29) Line412, Figure 20 should be the city.
30) Line408-Line420, there is a phenomenon of duplicate imaging in the image. How to remove it? As shown in the following figure
31) Line 471, Figure 23, what does 10 block mean? Where are the 10 blocks from? What is the method of correction of vignetting of multiple CCDS, how to implement this method, which was not previously introduced.
32) Line473, whether the RA, SM, and RSVN values in Table 3 are the average of 10 scenes, please explain
33) Line 489, Figure 24 does not provide valid information, such as how many JL03D satellites there are, how many scene images each satellite has, and what type of ground object. It is recommended to turn it into a table.
34) Line491 and Figure 25 are all thumbnails without any comparison before and after correction, and there is no valid information. It is recommended to delete them.
35) Line492, Table 4, it is recommended to provide CAM values for the raw image, histogram matching method, and new method.
36) Line510, is the number of scenes or scene number in Table 5
Comments for author File: 
Comments.pdf
Author Response
Thank you for your review. Please see the attachment.
Author Response File: Author Response.docx
Round 2
Reviewer 2 Report
Comments and Suggestions for AuthorsThe article has undergone extensive revisions, and the overall quality has been significantly improved.However, there are still some minor issues that need to be addressed.
line 182，What is the width of Figure 2？
line 187，Since the width of the sensor is 400, why did you choose a width of 200 (as shown in line 193)?
line 246，Modifications are required for (a) and (b)
line 275 ，(a) and (b) need to be removed
line 353，What is the abbreviation of RA?
line 456 Table 2，Is maximum linearity or minimum linearity?
line 460 ，Why the height of the satellite image is 600?
Comments on the Quality of English Language
Minor editing of English language required
Author Response
Thank you for your review. Please see the attachment.
Author Response File: Author Response.docx
