SSML: Spectral-Spatial Mutual-Learning-Based Framework for Hyperspectral Pansharpening
Round 1
Reviewer 1 Report
This paper proposes a new framework for image processing based on neural networks. However, authors must make the objectives of the paper clear.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 2 Report
The submitted manuscript proposed an SSML framework integrating spectral-spatial informationmining for HSI pansharpening.Latest fusion results are used to verify the ability of the SSML framework, the result is also to achieve better results. Both the theoretical details and the results are well described in the manuscript.I am very happy to recommend it for publication.
This paper compared the quality indicator results of different methods on PSNR,CC,SAM,RMSE,ERGAS. However,I would recommend to reconsider application and analysis of the efficiency and STD(Standard Deviation) of this proposed SSML framework, the time cost is used to judge the the efficiency the method , and STD is used to quantify the amount of variation or dispersion of a set of data values.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 3 Report
1) A relevant description of the spectral and spatial network is required.
2) How does the variation in training samples affect the performance of the proposed model
method?
3) The mathematical expressions for evaluation metrics are required to be included.
4) The experiments do not demonstrates the hyper-parameters' analysis.
Author Response
Please see the attachment.
Author Response File: Author Response.docx