Hyperspectral Super-Resolution Reconstruction Network Based on Hybrid Convolution and Spectral Symmetry Preservation
Round 1
Reviewer 1 Report
The authors of this manuscript propose a a hybrid convolution and spectral symmetry preservation network for hyperspectral images. The carried out performance evaluation illustrates comparable or improved performance over counterpart algorithms from the literature. The introduction is well-written and paves the way for the reader to understand the context of this research area and its latest developments. The proposed network is well-described. This is aided with the provision of the related mathematical formulae and figures. The numerical results and related plots provide good analysis and commentary on the performance of the proposed network, as well as compares it to the literature. The conclusions summarize the main ideas of the manuscript and are based on the previous sections. The provided references are adequate and relatively recent.
However, I have the following comments, which I believe need to be addressed before this manuscript could be considered for publication:
1. A language proof-reading is needed to correct some grammar and spelling mistakes.
2. In line 171, the acronym MER is used. This is strange since no 3 consecutive words begin with the sequence of letters M, E, R.
3. This is repeated in line 177 for the acronym GV
4. Each of the images in Figure 7 should be provided at a larger size. Their current dimensions are too small for reading and visual examination.
A thorough English language proof-reading is required.
Author Response
Please see the attachment
Author Response File: 
Author Response.docx
Reviewer 2 Report
The paper presents a particular solution to hyperspectral super resolution ..
The abstract lacks result details.
Paper goals/aims/objectives are not explicitly stated ...
Res3D, ECA are used without prior definition.
The paper presents in a descriptive style of presentation a particular architecture/solution, which is not well motivated. There is no discussion on possible design options/alternatives, parameter choices (sensitivities) etc.
It is mentioned, that only limited data is available for the rather 'greedy' method, augmentation techniques are deemed necessary and are used.
There is no discussion or evaluation on possible (surprising) generalization issues of the computing network itself and the particular created training data sets.
The investigated data sets/applications seem also to be quite limited in scope and it does not become really, what benefit is achieved by this very particular methods. Leave alone the computational and development cost associated with it.
The no. of training data used cannot be found in the paper, neither parameter settings etc. Repetition of the work and the results seems to be not straight forward.
l136+ weird sentence, l152 ... Typo
lower case start in several figure captions
Author Response
Please see the attachment
Author Response File: Author Response.docx
Reviewer 3 Report
The research topic of this paper has some value. The reviewer has the following suggestions:
1. Abstract: It is suggested that the author quantitatively analyze the advantages of the proposed model.
2. introduction: The author should have a clear motivation. Specifically, they should explain clearly what problems each component is designed to solve.
3. The author should properly classify and summarize the related works. Specifically, authors may refer to the related works sections of the following papers:
[1] A new hybrid ensemble deep reinforcement learning model for wind speed short term forecasting
[2] Pre-training enhanced spatial-temporal graph neural network for multivariate time series forecasting
[3] Diffused convolutional neural network for hyperspectral image super-resolution
4. It is suggested to adjust the font of the figures in the experimental results section. Now a lot of the figures are not clear
5. It is recommended to add relevant references to the evaluation index
6. What are the limitations of the model? It is suggested that the authors add limitations and future work in Section 5 (Conclusion)
The reviewer advised the authors to check the papers carefully for grammatical errors
Author Response
Please see the attachment
Author Response File: Author Response.docx
