An Improved Method for Evaluating Image Sharpness Based on Edge Information

Round 1

Reviewer 1 Report

Brief summary

This study proposes an improved image sharpness evaluation method. Its main contribution consists in performing the evaluation without reference image based solely on information extracted from edges (Canny edge detection algorithm, eight-neighborhood grayscale difference method and comparing 3 distance factors based on the histogram of edge width).

 

Broad comments

The document is generally easy to read and follow.

The English needs some minor spell checking.

The document is well supported with references.

The subject of the paper has great potential of application.

 

One of the weaknesses of this study is that it does not present a comparative study of the total processing time needed for the proposed method and the two other methods.

 

 

Specific comments

 

Figure 3 should only be inserted after being referenced in the text. Please correct.

 

Figure 4 should identify clearly the image resulting from improved algorithm, low threshold and high threshold. Please add that information to the image.

 

Figure 7 should also be inserted after being referenced in the text. Please correct.

 

In line 226 please standardize the reference to figures. (Fig. Figure ?!)

 

Figure 13 and 14 are referenced in the text before Figure 12. Please correct.

 

In line 241 the sentence “…the sharpness evaluation index of this paper can be obtained by…” is not very clear. Please rephrase.

 

Figure 15 is presented before it is referenced in the text. Please correct.

 

Table 1 should be as close as possible to the text where it is referenced. Please correct.

 

Also Figure 20 should be inserted after being referenced in the text.

 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

The authors present a method to evaluate the sharpness of an image. This is performed by estimating the edges and evaluating the width of an edge around some random edge points. The smaller the width the sharper the image. Therefore, they evaluate the distribution of edge widths and deducing some ad hoc measurements from the distribution, they determine the sharpness of the image. 

The paper is written well, the content is good, the experiments validate the claims. They propose a fast method to evaluate image sharpness and thus determine the amount of blur that corrupts an image. The reviewer just believes that it is of limited interest to the image processing community nowadays.

However, I have to recommend its acceptance, since there is nothing else to report.

 

 

Author Response

We are grateful for your review comments.

Reviewer 3 Report

Current form of manuscript not suitable to publish due to the following concerns,

1. Novelty is low

2. LAck of background study 

3. Lack of experimental analysis: Authors just reported the result

4. Lack of technical details: Justification for the chosen techniques is missing.

5. Complexity analysis missing.

6. Authors should compare the performance of the proposed with recent image analysis methods to claim superiority .

7. I fail to find the author's unique contribution to the proposed method.

 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 3 Report

The authors addressed all my comments well. I recommend the revised version of the manuscript for possible publication.