Skip to main content

Table 4 The content-specific validation performance values for Dataset II.

From: A framework for measuring sharpness in natural images captured by digital cameras based on reference image and local areas

Content

Metric

ROCC

LCC

RMSE

RMSE (subjective)

  

Training

Validation

Training

Validation

Training

Validation

 

Content 1

Proposed metric

0.847

0.666

0.838

0.674

12.20

13.620

24.438

 

Wang [18]

0.847

-0.275

0.838

-0.413

12.20

32.590

 

Content 2

Proposed metric

0.841

0.820

0.846

0.817

11.098

17.952

24.845

 

Wang [18]

0.499

0.829

0.591

0.761

16.79

18.620

 

Content 3

Proposed metric

0.836

0.849

0.857

0.800

12.287

14.265

21.503

 

Wang [18]

0.563

0.845

0.598

0.822

17.768

14.422

 

Content 4

Proposed metric

0.792

0.886

0.804

0.890

13.071

10.295

21.435

 

Wang [18]

0.527

0.877

0.587

0.829

17.78

14.699

 

Content 5

Proposed metric

0.782

0.903

0.782

0.915

12.935

11.168

19.900

 

Wang [18]

0.449

0.896

0.537

0.883

17.488

15.642

 
  1. Boldface indicates the best performer.