Haze Level Evaluation Using Dark and Bright Channel Prior Information

Round 1

Reviewer 1 Report

Overall a very good and interesting manuscript. The authors have made some significant contribution and well explained. The manuscript has a good flow, explanation and very well elaborated with interesting figures that may capture the interest of the readers. For further improvement the authors may 

1) Improve on the format of the tables.

2) Figure 1, the authors indicated that for the bottom row the haze level remain the almost the same. From my point of view, I can't see why the figures look the same.  

 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Point1. The expression referring to the image in the text is shown in Fig. Figure is mixed. Matching them together seems to be consistent.

Point2. Equation (4) of 2.2 Dark Channel Prior is the same as the equation cited from the paper [28] "Single Image Haze Removal Using Dark Channel Prior". The meaning is the same, but why change the order of min?

Point3. It seems to be clearer to specify Figure 3. (c), (f) in the explanation at the end of the lines (157~161) of the Bright Channel prior.

Point4. Tables 2 and 3 have the same caption. It seems that the caption of Table 3 is misspelled.

Point5. 3.1 There are grammatical incorrect or informal expressions.

    ex: Motivation, line 173, "why not we change ..." is grammatically incorrect

    ex: The beginning of line 391 of 4.4 Performance Analysis is "But anyway..."

Point6. Why is the hit count set to top 3 instead of top 1 in the result comparison table?

Point7. 4.6. In the case of Computation Cost, why did we test with one photo of Beijing rather than the average of multiple photos? And the results show that DBCP is not cost effective compared to other methods.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

All of the raised issues were adequately addressed.