Skip to main content

Table 3 Comparison of the performance of the proposed and the other algorithms against DIBCO dataset

From: An innovative document image binarization approach driven by the non-local p-Laplacian

Dataset Models FM % Fps % PSNR DRD
DIBCO 2009 Wang [10] 77.96 81.42 14.77 14.71
Jacobs et al. [11] 75.19 77.63 15.04 12.50
Jacobs et al. [12] 75.71 76.46 13.74 7.46
Rivest-Hénault et al. [15] 75.50 75.30 14.37 19.64
Guo et al. [14] 83.31 84.92 16.64 9.40
Proposed 88.34 93.03 17.41 4.98
DIBCO 2010 Wang [10] 77.63 86.13 16.26 5.60
Jacobs et al. [11] 49.65 54.84 14.35 9.75
Jacobs et al. [12]. 67.62 68.78 14.66 8.02
Rivest-Hénault et al. [15] 69.97 76.41 15.37 7.45
Guo et al. [14] 86.75 89.74 17.93 3.62
Proposed 88.80 94.46 18.60 3.06
DIBCO 2011 Wang [10] 81.94 86.98 15.76 7.26
Jacobs et al. [11] 73.33 77.26 14.32 13.92
Jacobs et al. [12] 77.90 74.60 14.68 6.28
Rivest-Hénault et al. [15] 69.40 70.19 12.70 51.25
Guo et al. [14] 83.82 87.32 16.50 6.55
Proposed 88.63 94.55 17.56 3.55
DIBCO 2012 Wang [10] 78.22 83.34 16.25 7.35
Jacobs et al. [11] 65.31 69.20 15.28 8.83
Jacobs et al. [12] 81.57 81.45 16.26 5.98
Rivest-Hénault et al. [15] 73.59 76.53 15.29 12.37
Guo et al. [14] 86.40 89.00 17.86 4.67
Proposed 88.89 93.41 18.81 3.74
DIBCO 2013 Wang [10] 80.07 83.98 16.51 9.51
Jacobs et al. [11] 73.66 77.24 15.98 9.85
Jacobs et al. [12] 77.72 82.34 16.54 9.20
Rivest-Hénault et al. [15] 77.33 78.86 15.81 11.48
Guo et al. [14] 82.35 85.16 17.37 8.09
Proposed 89.62 94.75 19.21 3.11
DIBCO 2014 Wang [10] 79.93 84.78 16.31 6.34
Jacobs et al. [11] 65.43 73.70 14.15 9.24
Jacobs et al. [12] 74.23 81.29 15.41 7.44
Rivest-Hénault et al. [15] 81.20 85.82 17.00 5.58
Guo et al. [14] 92.30 94.86 19.17 2.37
Proposed 92.34 96.03 19.02 2.55
DIBCO 2016 Wang [10] 86.92 89.93 18.05 4.61
Jacobs et al. [11] 79.09 81.13 16.67 6.13
Jacobs et al. [12] 83.25 87.98 17.66 5.32
Rivest-Hénault et al. [15] 83.06 84.43 16.49 7.27
Guo et al. [14] 88.51 90.46 18.42 4.13
Proposed 90.22 93.82 18.94 3.60