An enhanced binarization framework for degraded historical document images

EURASIP Journal on Image and Video Processing

Table 1 Performance evaluation results of our proposed method against the TOP 3 winners of the annual DIBCO or H-DIBCO competitions (best results highlighted in bold)

Dataset	Method	FM (%)	pFM (%)	PSNR (dB)	NRM (%)	DRD	MPM (‰)
DIBCO 2009	Rank 1st [21]	91.24		18.66	4.31		0.55
	Rank 2nd [55]	90.06		18.23	4.75		0.89
	Rank 3rd [56]	89.34		17.79	5.32		1.90
	Proposed	93.46		20.01	2.59		1.54
H-DIBCO 2010	Joint 1st [23]	91.50	93.58	19.78	5.98		0.49
	Joint 1st [57]	89.70	95.15	19.15	8.18		0.29
	Rank 2nd [58]	91.78	94.43	19.67	4.77		1.33
	Rank 3rd [28]	89.73	90.11	18.90	5.78		0.41
	Proposed	93.73	95.18	20.97	3.64		0.29
DIBCO 2011	Rank 1st [59]	80.86		16.13		104.48	64.43
	Rank 2nd [24]	85.20		17.16		15.66	9.07
	Rank 3rd [28]	88.74		17.84		5.36	8.68
	Proposed	90.72		18.85		4.47	7.87
H-DIBCO 2012	Rank 1st [29]	89.47	90.18	21.80		3.44
	Rank 2nd [59]	92.85	93.34	20.57		2.66
	Rank 3rd [24]	91.54	93.30	20.14		3.05
	Proposed	94.26	95.16	21.68		2.08
DIBCO 2013	Rank 1st [24]	92.12	94.19	20.68		3.10
	Rank 2nd [29]	92.70	93.19	21.29		3.18
	Rank 3rd [60]	91.81	92.67	20.68		4.02
	Proposed	93.51	94.54	21.32		2.77
H-DIBCO 2014	Rank 1st [33]	96.88	97.65	22.66		0.90
	Rank 2nd [29]	96.63	97.46	22.40		1.00
	Rank 3rd [61]	93.35	96.05	19.45		2.19
	Proposed	96.77	97.73	22.47		0.95
H-DIBCO 2016	Rank 1st [34]	87.61	91.28	18.11		5.21
	Rank 2nd [62, 63]	88.72	91.84	18.45		3.86
	Rank 3rd [62]	88.47	91.71	18.29		3.93
	Proposed	89.64	93.56	18.69		4.03
DIBCO 2017	Rank 1st [45]	91.04	92.86	18.28		3.40
	Rank 2nd	89.67	91.03	17.58		4.35
	Rank 3rd [42]	89.42	91.52	17.61		3.56
	Proposed	89.37	90.80	17.99		5.51
H-DIBCO 2018	Rank 1st [64]	88.34	90.24	19.11		4.92
	Rank 2nd	73.45	75.94	14.62		26.24
	Rank 3rd	70.01	74.68	13.58		17.45
	Proposed	88.34	90.37	19.11		4.93