A Multicriteria Evaluation of Single Underwater Image Improvement Algorithms
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsHigh quality underwater images are the basis of ecological monitoring, and the evaluation of existing image improvement algorithms lacks consideration of computational efficiency. This paper proposes a method based on multi criteria decision analysis to screen the preprocessing algorithm suitable for coral reef monitoring from the monocular underwater image database provided by veracruzana University, and comprehensively evaluate the processing time of the algorithm and the increase of feature points. Suggestions: 1. Strengthen the verification of water depth performance below 5 meters. 2 explore different algorithm combination strategies to optimize the map strength processing process.
Author Response
Comments 1: Strengthen the verification of water depth performance below 5 meters.
Response 1: We appreciate you pointing that out. ICIMAP database lacks video and
image depth information; however, we split single image by channels in the RGB model,
visualizing the red channel as black. In the state-of-the-art, this phenomenon occurs in
underwater imaging from 5m depth in turbid water and 20 m in clear water. We have
added the lines 297 – 303 to emphasize this point in page 12, paragraph 6.
Comments 2: Explore different algorithm combination strategies to optimize the map
strength processing process.
Response 2: Thank you for your observation. We have addressed this work in proposing
a framework to evaluate improvement algorithms that includes processing time as a
selection criteria, since in underwater ecological monitoring from imaging involves
several challenges. First challenge we are attending is to improve underwater image as
preprocessing stage to create a photomosaic. By combining methods in preprocessing,
processing time will increase and impact in computational efficiency.
Author Response File: Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsThis manuscript proposes a novel method built on multicriteria decision analysis that evaluates the processing time and the feature points increase with respect to the original image. The following issues are present in the manuscript:
1. Although many advanced underwater image enhancement and restoration methods have been proposed in recent years, why does the manuscript choose to compare with relatively older methods such as Sea-thru, DCP, and CBFWB? Would the inclusion of more recent state-of-the-art methods potentially lead to a more convincing evaluation of the proposed approach?
2. In the experimental section, a more comprehensive demonstration of the method’s performance would be achieved by presenting quantitative results averaged across the entire dataset, rather than reporting metrics derived from a single image.
3. There has been a growing body of research in the field of underwater image enhancement, but several key studies have been overlooked. Notably, recent advancements such as underwater color disparities: cues for enhancing underwater images toward natural color consistencies, inspiration: a reinforcement learning-based human visual perception-driven image enhancement paradigm, etc.,, to provide a more comprehensive overview of the field.
4. An increase in the number of keypoints is sometimes caused by the introduction of additional noise in the image. How does the manuscript ensure that the increase in keypoints produced by the proposed method corresponds to meaningful and valid features, rather than being a result of noise amplification?
5. Discussion and analysis of the failure cases should be provided for a better understanding of the proposed method.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for AuthorsThe present manuscript presents a multicriteria evaluation of single underwater image improvement algorithms. The analysis lacks the innovative character that a journal as JMSE requires.
The topic is of no relevance to the scientific community. As for the multi-criteria analysis, an analysis using only two criteria and three alternatives is very simple.
To a lesser extent, other comments:
- Please review the affiliations, I am referring to the sentence “Current address: Affiliation”.
- Lines 69, 72, 73, 75, 78, etc. Please use an impersonal form. Terms such “our” and “we” should be avoided in a scientific article.
- Line 114, please change “dehazing-based. [16]” to “dehazing-based [16].”. Idem n line 120.
- Section 2 (Background knowledge) should be merged with Section 1 (Introduction).
- Section 3 (Materials and Methods) is too short.
- Section 4 (Results and discussion) is very short. It should be expanded with a broader and more critical explanation of the results obtained.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
Comments and Suggestions for AuthorsThe authors have well addressed all my concerns in the revision. The manuscript is acceptable for publication in its present form.
Reviewer 3 Report
Comments and Suggestions for AuthorsThe authors have appropriately addressed my suggestions.