Color Filter Array Interpolation based on LPA-ICI
for noiseless and noisy CCD/CMOS sensor data

Software

The LASIP routines are available free-of-charge for educational and non-profit scientific research, enabling others researchers to understand and reproduce our work.

Any unauthorized use of the LASIP routines for industrial or profit-oriented activities is expressively prohibited. Please read the LASIP limited license before you proceed with downloading the files.

CLICK HERE TO DOWNLOAD .ZIP PACKAGE

IMPORTANT: This software is based (and requires) the LASIP image restoration demobox.

Table 1: PSNR comparison for different demosaicing methods computed excluding 15 pixels border: HA [1], LI [2], HD [3], SA [4], DFPD [5], AP [6], CCA [7], CCA+PP is a demosaicing approach [7] with postprocessing [8], DLMMSE [9], proposed LPA-ICI based interpolation in [10], "Oracle Γ" is the proposed LPA-ICI based interpolation [10] with the optimal threshold parameter Γ. It can be reproduced with demo_cfai_lpaici.m in the provided MATLAB codes.

			HA [1]	LI [2]	HD [3]	SA [4]	DFPD [5]	AP [6]	CCA [7]	CCA+PP [8]	DLMMSE [9]	LPA-ICI [10]	Oracle Γ [10]
1		R G B	33.17 34.65 33.29	30.87 35.61 30.98	34.51 36.13 34.74	36.99 40.76 38.77	36.15 38.14 36.45	36.69 40.42 37.26	36.22 39.20 36.72	37.25 41.31 38.29	37.58 40.22 38.02	39.49 42.51 39.95	39.91 43.11 40.48
2		R G B	37.38 40.94 39.60	36.54 41.35 37.32	36.83 41.61 40.44	35.50 40.57 40.05	38.17 43.09 41.34	37.29 42.46 40.84	37.76 43.81 41.35	36.52 43.27 41.06	38.19 44.31 42.54	38.78 44.63 42.79	38.72 44.64 42.81
3		R G B	40.21 42.19 39.79	39.22 43.15 38.37	40.39 43.56 40.03	39.19 41.00 38.84	41.41 44.75 40.83	40.94 43.52 40.34	41.58 45.18 41.20	39.90 43.78 40.30	41.95 45.80 41.40	42.90 46.06 42.29	42.91 46.05 42.30
4		R G B	36.58 40.66 39.64	36.73 42.32 38.91	35.85 41.47 40.74	35.25 41.63 41.91	37.01 43.12 41.93	36.87 43.81 42.33	37.40 44.55 42.46	35.92 44.26 42.32	37.21 44.68 43.61	37.77 44.53 43.42	37.87 44.55 43.38
5		R G B	34.65 35.81 34.27	32.60 36.75 32.25	34.95 37.27 34.27	34.60 36.78 34.26	36.87 39.45 36.18	36.87 39.69 36.06	37.12 40.02 36.42	36.10 39.61 35.72	37.62 40.88 36.71	37.03 40.03 36.29	38.06 41.05 37.17
6		R G B	34.66 35.93 34.24	32.49 36.91 31.98	37.35 38.88 36.49	39.02 42.22 38.00	39.05 40.81 38.00	38.22 41.48 37.35	36.86 39.86 36.37	38.12 41.68 37.43	40.13 42.33 38.80	40.86 43.51 39.34	40.93 43.76 39.33
7		R G B	40.64 42.17 40.25	38.77 42.16 38.51	39.99 42.64 39.37	39.25 41.22 38.75	41.26 43.98 40.56	41.25 43.96 40.69	41.87 45.31 41.32	40.07 43.97 39.91	41.83 45.27 41.01	42.64 45.50 41.82	42.68 45.52 41.86
8		R G B	31.57 33.37 31.55	28.05 32.96 27.83	33.06 35.19 33.11	34.55 38.45 35.28	34.51 37.17 34.63	34.56 38.55 34.67	33.34 36.93 33.46	34.44 38.63 34.82	35.08 38.53 35.23	36.10 40.00 36.26	36.08 40.07 36.32
9		R G B	39.38 41.39 40.23	36.78 41.40 37.39	40.12 42.83 40.45	39.72 42.02 40.90	41.02 44.18 42.78	40.64 43.42 41.90	40.99 43.94 40.92	40.29 43.90 40.91	41.69 45.14 43.00	42.31 45.44 43.30	42.30 45.42 43.31
10		R G B	39.12 41.33 39.49	37.64 42.27 37.69	39.43 42.80 39.89	40.25 43.91 41.45	40.79 44.38 41.75	40.78 44.38 41.41	40.83 44.42 40.96	39.94 44.41 40.67	41.19 45.37 42.08	41.63 45.47 42.10	41.76 45.53 42.19
11		R G B	35.50 37.03 35.87	33.71 37.90 33.65	36.44 38.76 37.25	37.43 41.20 39.03	38.11 40.67 38.83	37.96 41.64 38.92	37.71 41.25 38.70	37.69 42.15 39.32	38.75 42.03 39.87	39.21 42.74 40.27	38.97 42.81 40.16
12		R G B	39.89 42.27 40.21	37.44 42.41 37.74	40.45 43.82 41.11	40.89 44.38 42.21	41.80 45.23 42.43	41.41 45.22 42.09	40.97 44.74 41.31	40.45 44.82 41.46	42.09 46.30 42.98	42.86 46.63 43.38	42.87 46.64 43.41
13		R G B	29.53 30.58 29.10	28.86 32.53 28.36	31.32 32.18 30.43	36.00 38.38 34.05	33.15 34.26 32.19	34.04 36.83 32.86	34.07 36.02 32.97	35.84 38.13 34.00	34.98 36.09 33.56	36.48 38.14 34.40	36.48 38.14 34.37
14		R G B	34.81 37.14 35.02	33.74 37.72 33.31	33.82 37.73 34.62	31.59 34.98 32.64	35.39 39.23 35.75	34.57 38.04 35.11	35.68 40.25 36.42	33.77 39.00 35.11	35.53 40.28 36.25	35.54 39.74 36.33	36.53 40.55 37.03
15		R G B	36.08 39.51 37.96	36.50 41.56 37.57	35.69 40.62 38.82	35.68 40.59 39.78	37.03 42.05 40.02	36.79 42.29 40.22	36.95 42.79 40.59	36.10 42.32 40.62	37.22 43.23 41.24	37.71 43.11 41.19	37.74 43.13 41.18
16		R G B	38.09 39.55 37.88	35.44 40.01 35.30	41.20 42.66 40.41	42.11 45.46 41.08	42.69 44.37 41.81	41.56 44.82 40.85	39.73 42.87 39.49	41.14 44.60 40.49	43.60 45.75 42.49	43.81 46.15 42.73	44.23 46.83 42.84
17		R G B	38.41 39.19 37.74	36.95 40.38 36.67	38.93 40.36 38.32	40.88 43.17 40.52	40.17 42.02 40.03	40.79 43.03 40.31	40.77 42.94 39.92	40.06 43.22 39.80	41.38 43.15 40.83	41.41 43.39 40.84	41.38 43.41 40.81
18		R G B	33.91 35.24 33.75	33.40 36.92 33.00	34.21 36.12 34.26	35.32 38.36 36.41	35.68 38.12 36.29	36.23 39.46 36.86	36.56 39.42 36.71	35.90 39.73 36.88	36.69 39.41 37.27	36.56 39.53 37.18	36.75 39.54 37.23
19		R G B	36.67 38.34 37.09	32.61 37.24 32.64	37.60 39.51 37.86	38.67 42.07 40.17	38.78 41.28 40.01	38.70 42.12 39.80	37.90 41.08 38.16	38.41 42.25 38.85	39.83 42.77 40.93	40.32 43.36 41.48	40.32 43.53 41.57
20		R G B	38.75 39.82 37.31	36.98 40.57 35.72	39.29 40.85 37.55	40.54 42.79 38.12	40.50 42.40 38.56	40.96 43.50 38.71	41.17 43.76 39.24	40.66 43.87 38.98	41.80 43.86 39.27	41.90 44.01 39.62	41.97 44.32 39.68
21		R G B	35.04 36.29 34.48	33.40 37.47 32.74	36.46 37.76 35.41	38.86 41.92 37.50	37.61 39.47 36.67	38.47 41.57 37.19	38.31 40.90 37.16	39.14 42.32 37.70	39.14 41.22 37.65	39.68 41.97 38.03	40.00 42.45 38.09
22		R G B	35.80 37.72 35.60	34.90 38.44 33.84	35.41 38.22 35.48	36.53 38.69 36.18	36.73 39.55 36.75	37.03 39.72 36.71	37.22 40.58 37.13	36.06 39.88 36.55	37.60 40.86 37.42	37.64 40.83 37.55	37.73 40.91 37.67
23		R G B	41.06 43.30 41.54	39.83 43.71 40.27	40.31 43.92 41.15	38.50 41.44 39.51	41.26 44.90 41.98	40.76 44.03 41.61	41.48 45.60 42.40	39.25 44.03 41.02	41.78 46.24 42.78	42.76 46.39 43.37	42.78 46.39 43.39
24		R G B	32.63 33.59 30.76	32.24 35.61 30.47	32.70 35.15 31.90	34.70 37.38 33.03	34.43 36.74 32.83	34.94 37.49 32.93	34.50 37.45 32.95	33.53 37.41 32.82	35.94 38.01 33.74	35.62 38.12 33.66	35.62 38.14 33.67
Mean PSNR		R G B	36.40 38.25 36.53	35.82 39.06 34.69	36.93 39.58 37.26	37.58 40.81 38.27	38.32 41.22 38.69	38.26 41.73 38.63	38.21 41.79 38.51	37.77 42.02 38.54	39.11 42.57 39.53	39.63 43.00 39.90	39.77 43.19 40.01

[1]. J.F. Hamilton Jr. and J.E. Adams, Adaptive color plane interpolation in single color electronic camera, U.S. Patent 5 629 734, May 1997.
[2]. H.S. Malvar, L.-W. He, and R. Cutler, High-quality linear interpolation for demosaicing of Bayer-patterned color images, IEEE Int Conf (ICASSP '04), Proceedings on Acoustics, Speech, and Signal Processing 3, 2004, 485-8.
[3]. K. Hirakawa, T.W. Parks, Adaptive homogeneity-directed demosaicing algorithm, IEEE Trans on Image Processing 14, Issue 3 (2005), 360-369.
[4]. X. Li, Demosaicing by successive approximation, IEEE Trans on Image Processing 14, Issue 3 (2005), 370-379.
[5]. D. Menon, S. Andriani, G. Calvagno, Demosaicing with directional filtering and a posteriory decision, IEEE Trans on Image Processing 16, No. 1 (2007), 132-141.
[6]. B.K. Gunturk, Y. Altunbasak, R.M. Mersereau, Color plane interpolation using alternating projections, IEEE Trans on Image Processing 11, Issue 9 (2002), 997-1013.
[7]. R. Lukac, K.N. Plataniotis, D. Hatzinakos, M. Aleksic, A novel cost effective demosaicing approach, IEEE Trans on Consumer Electronics 50, No. 1 (2004), 256-261.
[8]. R. Lukac, K. Martin, K.N. Plataniotis, Demosaicked image postprocessing using local color ratios, IEEE Trans on Circuits and Systems for Video Technology 14, No. 6 (2004), 914-920.
[9]. L. Zhang, X. Wu, Color demosaicking via directional linear minimum mean square-error estimation, IEEE Trans on Image Processing 14, No. 12 (2005), 2167-2178.
[10].D. Paliy, V. Katkovnik, R. Bilcu, S. Alenius, K. Egiazarian, "Spatially Adaptive Color Filter Array Interpolation for Noiseless and Noisy Data", Int. J. Imaging Sys. Tech., Sp. Iss. Appl. Color Image Process., vol. 17, no. 3, pp. 105-122, Oct 2007.

Table 2: PSNR comparison: GAUSSIAN noise (σ=12.75).

Noisy Bayer Data		7	8	13	19	23	Mean PSNR
HA (Hamilton, 1997) (illustrates noisyness of RGB)		25.79 26.47 25.88	24.86 25.69 24.77	24.36 25.07 24.26	25.55 26.24 25.57	25.91 26.54 25.95	25.29 26.00 25.29
HA (Hamilton, 1997) with prefiltering (Katkovnik, 2002)	R G B	30.29 31.06 29.60	26.50 27.30 26.17	24.56 24.79 24.69	29.11 29.94 29.34	31.93 32.66 30.77	28.48 29.15 28.11
AP (Gunturk, 2002) with prefiltering (Katkovnik, 2002)	R G B	30.51 31.06 30.53	27.24 27.97 27.39	25.25 25.37 25.65	29.35 30.11 30.09	31.96 32.60 31.93	28.86 29.42 29.12
DLMMSE (Zhang, 2005) with prefiltering (Katkovnik, 2002)	R G B	30.89 31.48 31.00	27.51 28.19 27.65	25.40 25.39 25.75	29.66 30.36 30.36	32.34 33.08 32.39	29.16 29.70 29.43
Proposed LPA-ICI based integrated demosaicing with denoising	R G B	32.12 32.40 31.94	27.29 28.09 27.59	25.61 25.45 25.96	29.74 30.59 30.61	33.18 34.17 33.34	29.59 30.14 29.89

Table 3: PSNR comparison: POISSONIAN noise (χ=0.5447).

Noisy Bayer Data		7	8	13	19	23	Mean PSNR
HA (Hamilton, 1997) (illustrates noisyness of RGB)		24.80 25.56 25.31	23.91 24.77 23.81	23.85 24.60 24.02	24.54 25.20 24.71	24.96 25.67 25.73	24.41 25.16 24.71
HA (Hamilton, 1997) with prefiltering (Foi, 2005)	R G B	30.07 31.01 29.87	25.72 26.61 25.42	25.00 25.60 24.93	28.17 29.07 28.61	31.59 32.57 30.79	28.11 28.96 27.92
AP (Gunturk, 2002) with prefiltering (Foi, 2005)	R G B	30.42 31.13 30.77	26.49 27.16 26.56	26.07 26.55 26.26	28.53 29.33 29.30	31.85 32.56 31.85	28.67 29.34 28.94
DLMMSE (Zhang, 2005) with prefiltering (Foi, 2005)	R G B	30.75 31.53 31.29	26.67 27.45 26.81	26.34 26.66 26.52	28.68 29.52 29.59	32.11 33.02 32.49	28.91 29.63 29.33
Proposed LPA-ICI based integrated demosaicing with denoising	R G B	31.74 32.30 32.02	27.15 27.92 27.29	27.21 27.30 27.12	29.42 30.16 30.16	32.56 33.45 32.83	29.62 30.22 29.88

Publications

2008

Paliy, D., A.Foi, R. Bilcu, V. Katkovnik, “Denoising and Interpolation of Noisy Bayer Data with Adaptive Cross-Color Filters”, SPIE-IS&T Electronic Imaging, Visual Communications and Image Processing 2008, vol. 6822, San Jose, CA, January 2008.

2007

Paliy, D., V. Katkovnik, R. Bilcu, S. Alenius, K. Egiazarian, “Spatially Adaptive Color Filter Array Interpolation for Noiseless and Noisy Data”, International Journal of Imaging Systems and Technology (IJISP), Special Issue on Applied Color Image Processing, vol. 17, iss. 3, pp. 105-122, October 2007.
Paliy D., R. Bilcu, V. Katkovnik, M. Vehvilainen, “Color Filter Array Interpolation Based on Spatial Adaptivity”, Proc. of SPIE-IS&T Electronic Imaging 2007, Computational Imaging IV, Vol. 6497, 649707, San Jose, CA, January 2007.
Paliy D., M. Trimeche, V. Katkovnik, S. Alenius, “Demosaicing of Noisy Data: Spatially Adaptive Approach”, Proc. of SPIE-IS&T Electronic Imaging 2007, Computational Imaging IV, Vol. 6497, 649720, San Jose, CA, January 2007.

---

Contacts

For any questions, suggestions, recommendations, corrections, contact D.Paliy (dmitriy.paliy(at)tut.fi)
back to top of page

---

Tampere University of Technology - Department of Signal Processing - Transforms and Spectral Methods Group