Steven S. O. Choy

Reduction of block-transform image coding artifacts by using local statistics of transform coefficients

This letter presents a new approach to reduce coding artifacts in transform image coding. We approach the problem in an estimation of each transform coefficient from its quantized version with its local mean and variance. The proposed method can significantly reduce coding artifacts of low bit-rate coded images, and at the same time guarantee that the resulting images satisfies the quantization error constraint.

New single-pass algorithm for parallel thinning

It is well known that many proposed parallel thinning algorithms cannot satisfy all major thinning requirements. In this paper we propose a new parallel thinning algorithm which can satisfy all major thinning requirements. The algorithm we present is a single-pass parallel thinning algorithm using reduction operators with 13-pixel support. A systematic derivation of the template set for the proposed algorithm is described. The proposed algorithm always requires a small number of iterations in thinning while at the same time it produces perfectly 8-connected medial curves. The proposed algorithm is evaluated and compared with other existing parallel thinning algorithms. It is shown from detailed experimental results that the new algorithm is superior to other algorithms in computation time and thinning results.

Adaptive image noise filtering using transform domain local statistics

Image noise filtering has been widely perceived as an estimation problem in the spatial domain. We deal with it as an estimation problem in an uncorrelated transform domain. This idea leads to a generalization of the adaptive linear minimum mean square error (LMMSE) estimator for filtering noisy images. In our proposed method, the transform-domain local statistics obtained from the noisy image are exploited. Due to the fact that the transform-domain local statistics carry more information about the image than the spatial-domain local statistics do, improvement in noise filtering is gained overall and is particularly significant in the vicinity of edges.

Reduction of coding artifacts in transform image coding by using local statistics of transform coefficients

This paper proposes a new approach to reduce coding artifacts in transform image coding. We approach the problem in an estimation of each transform coefficient from the quantized data by using its local mean and variance. The proposed method can greatly reduce coding artifacts of low bit-rate coded images, and at the same time guarantee that the resulting image satisfies the quantization error constraint.

New image restoration performance measures with high precision

The field of image restoration lacks a promising comparison vehicle for judging the effectiveness of competing algorithms. By far the most widely adopted quantitative measurement of image restoration per- formance is by means of the SNR improvement. We address the issues of performance assessment of image restoration and propose a unified framework for the performance measure. The SNR improvement, which is within the proposed framework, is shown to be an inappropriate per- formance measure for image restoration. By introducing a metric for pixel fidelity improvement and incorporating the main properties of the human visual system into the measurement, we devise a performance measure of better quality, particularly of higher precision.

New adaptive iterative image restoration algorithm

It has been shown in the literature that adaptive regularized image restoration is superior to the non-adaptive case. However, the adaptivity introduced in most proposed iterative algorithms is based only on the application of the space-variant smoothing operator. It is found that these adaptive algorithms suffer from insufficient smoothing of the flat image regions. In this paper, an adaptive iterative image restoration algorithm, which applies both techniques of space-variant smoothing and space-variant restoration, is proposed to overcome the stated problem. It is shown by experiments that the restored images obtained by the proposed algorithm are better in terms of both numerical measurement and visual quality.<<ETX>>

Regularized restoration of VQ compressed images with constrained least squares approach

In this paper, an iterative algorithm is proposed to restore VQ-encoded images. This algorithm incorporates adaptivity into simple CLS restoration algorithm by weighting every pixel according to its expected derivation from the original. This algorithm is fully compatible with any VQ codec to improve the codecs coding performance. Computer simulations showed that the proposed scheme was superior to other conventional schemes in terms of SNR improvement. Besides, the image quality could also be improved subjectively by reducing the blocking effect.

A new approach for restoring block-transform coded images with estimation of correlation matrices

This paper presents a new restoration approach to reduce coding artifacts in block-transform image coding. Different from conventional restoration techniques, the proposed one is non-iterative and requires a low computational cost, yet can reconstruct objectively and subjectively better images. This good performance is achieved because of the following advantages the proposed approach has: (i) efficient incorporation of the solution bound into restoration; and (ii) effective exploitation of local image properties and statistical knowledge about the quantizers used.

Optimal choice of local regularization weights in iterative image restoration

In the study of space-variant regularization for image restoration, little effort has been devoted to the search of optimal local regularization weights. In this paper, we address how to derive the optimal local regularization weights in the context of iterative image restoration. The optimal relationship between the two weight matrices for local regularization is derived, and, based on that relationship, a proper choice of the weight matrices is then presented. The results we derived provide a mathematical backup of the viability of some heuristic solutions suggested in the literature.