Hyperspectral Image Super-Resolution via Adaptive Factor Group Sparsity Regularization-Based Subspace Representation

Round 1

Reviewer 1 Report

This paper proposed a hyperspectral image superresolution via adaptive factor group sparsity in subspace. The method is demonstrated using experiments. The following questions can be considered before acceptance.

1. In modeling the problem, S is solved as difference map. Such modeling is not commen seen as far as I know, please discuss more on it.

2. The inferring time is fast, please explain more on it. Normally iterative method is time consuming.

3. More recent sparsity based method can be compared.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Kindly check in the attachment

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Kindly compare with more hyperspectral datasets.

Minor editing is required

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

-> The paper lacks clarity in describing the proposed Factor Group Sparsity Regularized Subspace Representation (FGSSR) method. The explanation of the FGSSR model's components and how they address specific challenges in hyperspectral image super-resolution is insufficient.

->  The paper lacks a detailed description of the experimental setup, including the hyperspectral datasets used, preprocessing steps, and parameter settings.

->  While the paper mentions the development of an "effective proximal alternating minimization-based algorithm" to solve the FGSSR-based model, there is no information on the algorithm's convergence properties, complexity analysis, or convergence criteria.

->  The paper mentions experimental results on simulated and real datasets, but it does not provide any qualitative or quantitative analysis of the results.

-> The paper lacks a discussion of the limitations of the proposed method. It is essential to acknowledge and address potential drawbacks or scenarios where the FGSSR method might not perform well.

->  Key concepts such as "Schatten-p norm" and "tensor nuclear norm regularization" need to be explained more thoroughly, as not all readers may be familiar with these terms.

-> Add future scope by considering  Evolving fusion-based visibility restoration model for hazy remote sensing images using dynamic differential evolution

 

Needs some major improvements.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 4 Report

I have no more suggestions