Multispectral image restoration with multisensors

Boo, K.J.; Bose, N.K.

doi:10.1109/36.628783

Cited by 24 publications

(15 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The multichannel (e.g., color) image restoration problem has recently attracted much attention in the imaging community (cf. [3,12,29,18,11]). In this paper, we study an alternating minimization algorithm for recovering multichannel images from their blurry and noisy observations.…”

mentioning

confidence: 99%

A Fast Algorithm for Edge-Preserving Variational Multichannel Image Restoration

Yang¹,

Yin²,

Zhang³

et al. 2009

SIAM J. Imaging Sci.

474

217

View full text Add to dashboard Cite

Abstract. We generalize the alternating minimization algorithm recently proposed in [32] to efficiently solve a general, edge-preserving, variational model for recovering multichannel images degraded by within-and cross-channel blurs, as well as additive Gaussian noise. This general model allows the use of localized weights and higher-order derivatives in regularization, and includes a multichannel extension of total variation (MTV) regularization as a special case. In the MTV case, we show that the model can be derived from an extended half-quadratic transform of Geman and Yang [14]. For color images with three channels and when applied to the MTV model (either locally weighted or not), the per-iteration computational complexity of this algorithm is dominated by nine fast Fourier transforms. We establish strong convergence results for the algorithm including finite convergence for some variables and fast q-linear convergence for the others. Numerical results on various types of blurs are presented to demonstrate the performance of our algorithm compared to that of the MATLAB deblurring functions. We also present experimental results on regularization models using weighted MTV and higher-order derivatives to demonstrate improvements in image quality provided by these models over the plain MTV model.

show abstract

mentioning

confidence: 99%

A Fast Algorithm for Edge-Preserving Variational Multichannel Image Restoration

Yang¹,

Yin²,

Zhang³

et al. 2009

SIAM J. Imaging Sci.

474

217

View full text Add to dashboard Cite

show abstract

“…Some works oriented towards remote sensing can be cited: Fonseca et al (1993) developed the Modified Inverse Filter, which is a regularized version of the Inverse Filter to restore and interpolate Landsat images; Wu and Schowengerdt (1993) dealt with the restoration of images containing mixed pixels; Bhaskar et al (1994) tackled the problem of lens defocus and linear motion blur in Space Shuttle images; Reichenbach et al (1995) described the design of small convolution kernels for the restoration and reconstruction of Advanced Very High Resolution Radiometer (AVHRR) images; Boo and Bose (1997) applied image restoration techniques to multispectral images; Jalobeanu et al (1993) used complex wavelet packets to deconvolve degraded images; Likhterov and Kopeika (2002) dealt with the problem of vibration in images with a differential scheme; Chen and Xanju (1994) The POCS method uses a priori knowledge about the image or the imaging system. The key to effectively apply this kind of algorithm is to define the appropriate sets, compute the projection onto these sets, and incorporate the projectors into an image processing algorithm designed to meet some criteria implied by the constraints (Stark 1998).…”

Section: Introductionmentioning

confidence: 99%

Convex restriction sets for CBERS‐2 satellite image restoration

Papa

Mascarenhas

Fonseca

et al. 2008

International Journal of Remote Sensing

View full text Add to dashboard Cite

The main goal of this work is to develop a new efficient image restoration algorithm based on projections onto convex sets technique (POCS), which uses some a priori information about the images in the form of restriction sets. The proposed convex formulation used in this work is the prototype image contraints, which was obtained by the Modified Inverse Filter, limited amplitude and by the Modified Row-Action Projection (MRAP) algorithm. A simulation experiment was performed using a high resolution IKONOS image, which was blurred and decimated according to CBERS-2 CCD camera specifications. In order to allow quantitative analysis, the ISNR and the Universal Image Quality Index methodologies were applied. An original CBERS-2 CCD image was also used to evaluate the proposed restoration method. The restored images show a good visual performance, which can also be observed by the autocorrelation coefficients, which indicate high frequency enhancement.

show abstract

“…This is especially true when the color-quantized version is required to be further processed or compressed. However, though there are a lot of reported works on the restoration of noisy and blurred color images [10][11][12][13][14][15][16][17][18], little effort has been seen in the literature for restoring halftoned color-quantized images. Obviously, the degradation models of the two cases are completely different and hence direct adoption of conventional restoration algorithms does not work effectively.…”

Section: Introductionmentioning

confidence: 99%

“…The number of input patterns fed into the LUT is expected to be very limited and hence it is possible to construct a LUT of a reasonably small size. In particular, for a 16-pixel neighborhood, it needs a LUT of only 64 2 16 K bytes. Besides, as there are only a few involved input patterns, just a few of training images are already good enough to provide sufficient number of training samples to construct the LUT.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Restoration of Halftoned Color-Quantized Images using Linear Estimator

Fung

Chan

2006

2006 IEEE International Conference on Multimedia and Expo

View full text Add to dashboard Cite

-Restoration of color-quantized images in which error diffusion is involved is rarely addressed in the literature. Conventional restoration algorithms were generally proposed for handling other noise models and hence they are not effective in restoring color-quantized images. In this paper, we tackle this problem by making use of the best linear estimation. It is shown by the simulation results that the algorithm can improve the quality of colorquantized images to a certain extent.

show abstract

Multispectral image restoration with multisensors

Cited by 24 publications

References 18 publications

A Fast Algorithm for Edge-Preserving Variational Multichannel Image Restoration

A Fast Algorithm for Edge-Preserving Variational Multichannel Image Restoration

Convex restriction sets for CBERS‐2 satellite image restoration

Restoration of Halftoned Color-Quantized Images using Linear Estimator

Contact Info

Product

Resources

About