Calibration of SAR Polarimetric Images by Covariance Matching Estimation Technique with Initial Search

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

In this work, the authors discuss a new method of calibrating polarimetric synthetic aperture radar (SAR) data. A brief introduction to the existing calibration algorithms is provided. The calibration model is defined and the various factors which affect the calibration are detailed.  The comet estimator is also introduced. The trouble and causes for improving the data are provided and the solution for obtaining better output using Comet IS is provided. The authors clearly describe the all the algorithms that are considered in this work. The new calibration method named “Comet with Initial Search (Comet IS)” is dependent on the Quegan and Comet algorithm outputs and is primarily based on excluding the outliers seen in the first pass of the calibration. The authors test the efficiency of the algorithms using signal simulations. The various variables considered for the simulations are described and the outputs of Ainsworth, Quegan, Comet and Comet IS are presented, and it is seen that the Comet IS improves the results compared to Comet algorithm. The results of testing the algorithms on real data is discussed, although visually it is challenging to get a sense of the advantages of Comet IS algorithm, the authors utilize the corner reflectors for validation and show that the output of Comet IS performs better than the other algorithms considered. The detailed evaluation metrics of all the corner reflectors are also provided. It is an interesting technique which improves the quality of SAR data, but it will be curious to see the computational performance of the proposed algorithm.  The authors however mention it in the conclusion that the computational overhead is one of the problems at hand. The topic is interesting as it presents a method for improving polarimetric SAR calibration. I recommend accepting after the authors address minor comments below:

 

•              Consider expanding the introduction section by adding details to the scope of the problem at hand and expanding the importance of calibration of SAR images.

•              Line 112, it is mentioned that that set “h” as 1 during estimation and then correct it in the subsequent procedure by corner reflectors. Does this imply that the method applies to corner reflector targets only, how does this affect the calibration with distributed targets.

•              Consider making the figure labels and legend larger. Also include units for figure 4 if relevant.

 

•              In the simulations, 927 outlier sets and 5673 true data was considered. Considering a complex scenario when the number of outlier data is comparable to the number of true data, what results can be expected from the proposed algorithm? Do we get significantly improved results?

Comments on the Quality of English Language

 

Consider go through the entire paper and correcting for minor grammatical errors. Examples below:

Line 83: matric -> matrix

Line 630: co-pol?

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

In this paper, an improved covariance matching estimation technique (Comet IS) is proposed to calibrate polarimetric SAR data. However, before publication, there are some minor question and comments as follow:

1.  Introduction:

  1) The introduction lacks relevant introductions to the methods and conclusions proposed in this paper. More relevant introductions should be added, and the importance and significance of the proposed method should be further highlighted.

  2) There are many paragraphs in the introduction. It is suggested to reorganize them in combination with recommendation 1, so as to make them more logical and better support the content mentioned in this paper.

2.  Format:

  1) Pay attention to the location of all formulas, so that they meet the format requirements. In this paper, the position of the formula is relatively front. Please refer to the formula example 1 in the template, set the formula line as a regular text, and place the formula in the middle of each line.

  2) Line 144, confirming that the naming number of “3.1. Comet Estimator” is correct.

  3) Please change the Table to the standard three-line table format.

3.  Figures:

  1) In Figure 2, the name of the ordinate axis is not clear, please correct it. Figure 5 is the same.

  2) In Figure 7, please put the subgraph numbers (a), (b), (c) directly below the figure, and improve the clarity of the legend.

  3) Line 191, the overall name of the picture should be added before the subgraph name in Figure 1. Figures 2, 4, 6, 7, 8, 9, 10 are the same as above.

4.  References:

  Please refer to the template style to modify the reference format.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The manuscript presents an improved method for calibrating polarimetric synthetic aperture radar (PolSAR) images using a covariance matching estimation technique with initial search (Comet IS). While the method aims to address issues of ill-conditioning in the iterative process of existing methods, there are several technical flaws and areas that require further clarification and additional experiments.

 

1.  The manuscript states that practical applications of the covariance matching estimation technique reveal issues stemming from ill-conditioning due to the analytical solution in the iterative process. However, the explanation of how the proposed initial search method specifically addresses these ill-conditioning issues is insufficiently detailed. Further elaboration is needed on the mathematical and algorithmic steps taken to mitigate this problem.

 

2. The paper introduces an outlier detection mechanism based on the Quegan algorithm's results. However, the criteria for identifying outliers and the threshold settings are not clearly defined. More detailed explanations and justifications for these criteria are necessary to understand the robustness and effectiveness of the outlier detection mechanism.

 

3. The use of the interior point method for recalibration is an interesting approach. However, the manuscript lacks a comprehensive discussion on the convergence properties of the initial search method. There should be a detailed analysis of the convergence rate, potential for local minima, and the computational complexity of this approach compared to traditional methods.

 

4. The simulation experiments are stated to reveal that the improved algorithm outperforms the original. However, the experimental setup, including the parameters, number of simulations, and statistical analysis methods, are not adequately detailed. The results should be presented with more rigor, including confidence intervals and statistical significance tests to validate the performance improvements.

 

5. The manuscript compares the improved method with the Quegan and Ainsworth algorithms, demonstrating its superior performance in calibration. However, the comparisons lack depth in several areas:

     - Specific scenarios or datasets where the Comet IS algorithm performs better or worse than the other algorithms are not sufficiently discussed.

     - The manuscript should include a more detailed quantitative comparison, including metrics like crosstalk isolation, channel imbalance, and overall calibration accuracy.

 

6. The validation using real data and corner reflectors is a strong aspect of the manuscript. However, the description of the data acquisition process and the specific conditions under which the data was collected need to be more comprehensive. Details such as the environmental conditions, radar parameters, and any preprocessing steps should be included.

7.There is a lack of sensitivity analysis on the parameters used in the Comet IS algorithm. Understanding how variations in parameters affect the calibration results is crucial for assessing the robustness and reliability of the proposed method. A thorough sensitivity analysis should be included.

8.The manuscript briefly mentions the assumption that noise between channels is mutually independent and uses average noise power. However, the impact of different noise levels and types on the performance of the Comet IS algorithm is not thoroughly investigated. A detailed study on how the algorithm performs under varying noise conditions would strengthen the paper.

9.The computational complexity of the Comet IS algorithm compared to the Quegan and Ainsworth algorithms is not discussed. Providing an analysis of the computational requirements, including time complexity and resource usage, would be valuable for understanding the practical applicability of the proposed method.

10.The manuscript should include a discussion on the stability of the Comet IS algorithm. Specifically, it should address whether the algorithm consistently converges to a solution and how sensitive it is to initial conditions and parameter settings.

11.To enhance the literature review and provide a broader context for the research, I recommend including the following references:

1. "An advanced scheme for range ambiguity suppression of spaceborne SAR based on blind source separation"
2. "Polarimetric SAR imaging: theory and applications"

Comments on the Quality of English Language

1：Abstract: Contains several grammatical errors and awkward phrasings. For example:

"Various methods have been proposed in the literature" should be "Various methods have been proposed."

"which based on the Quegan algorithm's results" should be "which is based on the results of the Quegan algorithm."

"Simulation experiments reveals" should be "Simulation experiments reveal."

2：Introduction：Provides a good overview but has multiple grammatical issues:

"PolSAR technology holds significant potential and value in various applications, including unsupervised land cover segmentation, target detection and recognition, as well as estimation of parameters such as soil moisture and biomass." This sentence is lengthy and could be split for better readability.

"Among this, channel imbalance and crosstalk are mainly caused by the inconsistent channel performance and energy leakage of the polarimetric channels." This should be rephrased for clarity.

3：Page 1, Line 17: "an improved method，Comet IS, is introduced." Replace the comma with "called".

4：Page 1, Line 25: "significantly improved." Change to "significantly improve."

5：Page 2, Line 31: "PolSAR systems harness not only the power information from imagery but also the relative phase information between channels, which quantitatively reflects differences in target scattering characteristics." Simplify to "PolSAR systems use both power and relative phase information from imagery to reflect differences in target scattering characteristics."

6：Page 3, Line 46: "calibration is the task of estimating and correcting for imbalances and cross-talks between the channels delivering the different polarimetric responses." Simplify to "calibration estimates and corrects for imbalances and crosstalk between polarimetric channels."

7：The manuscript requires a thorough grammatical review to enhance clarity and readability. 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

Congratulations on the substantial improvements made to your manuscript. The concerns I previously raised have been largely addressed. I believe this version of the manuscript now meets the publication standards.