A new algorithm is presented for recovering the blocks lost with the cell loss in the ATM transmission of the images coded by Jacquin's fractal coding. The key technique of the proposed BLRA (block loss recovery algorithm) is a fractal extrapolation that estimates the pixels in a lost block by using the contractive mapping parameters of a range block homogeneous to the lost block. The proposed BLRA is applied to the lost blocks in the iteration of decoding.

An efficient algorithm is proposed for the adaptive sizing of a tracking window in correlation-based video tracking. Since a tracking window specifies a support region when estimating a target displacement, the ability to adapt the window size relative to a moving target significantly influences the performance of video tracking. The basic strategy of the proposed algorithm is to maintain the occupancy rate of the target in the tracking window within a specified range. As such, the proposed algorithm measures the occupancy rate using the ratio of the power of the spatial gradients in the edge subwindows, which edge the tracking window, to that in the tracking window. In addition, the level of any complex background and additive white noise is also evaluated to reduce their effect on the gradients. Experimental results using various artificial and real sequences confirm that the proposed algorithm can effectively adjust a tracking window to a moving target and is robust to a complex background and noise.

We present an efficient scheme for progressive fractal coding in a wavelet multiresolution pyramid. In our proposed scheme, range blocks are selectively coded, based on energy localization in wavelet bands. In order to form domain pools with blocks yielding less distortion, domain blocks are also selectively searched in the wavelet bands of either the same or a different orientation from that of each range block at the next lower resolution. It is shown that the performance of the non-hybrid fractal coder presented is much better than that of predictive pyramid coding (PPC).

In this paper, we propose a fast fractal image coding based on LMSE (least mean square error) analysis and subblock feature. The proposed method focuses on efficient search of contrast scaling, position of its matched domain block, and isometric transform for a range block. The contrast scaling and the domain block position are searched using a cost function that comes from the LMSE analysis of the range block and its fractal-approximated block. The isometric transform is searched using 2 2 blocks formed with the averages of subblocks of range block and domain block. Experimental results show that the encoding time of a conventional fractal image coding with our search method is 25.6-39.7 times faster than that with full search method at the same bit rate while giving PSNR decrement of 0.2-0.7 dB with negligible deterioration in subjective quality. It is also shown that the encoding time of a conventional fractal image coding with our search method is 3.4-4.2 times faster than Jacquin's fractal image coding and is superior by maximum 0.8 dB in PSNR. It also yields reconstructed images of better quality.

We present a model of quantization noise in block-coded videos with some assumptions in wavelet domain and propose a postprocessing method to reduce the quantization noise based on the model. A frame of video sequences is considered as a set of one-dimensional (1-D) horizontal and vertical signals. The quantization noise is considered as the sum of the blocking noise and the remainder noise. We model the blocking noise as an impulse or that along with a dispersed impulse at each block boundary in the wavelet domain. The validity of the blocking noise model is investigated. We also model the remainder noise as white Gaussian noise at non-edge pixels in the wavelet domain. Whether the model accommodates well to the remainder noise or not is also examined. The blocking noise is reduced by subtracting a profile, whose strength is adaptively estimated, at each block boundary from the coded signal. The remainder noise then is reduced by a soft-thresholding. We also propose a fast algorithm for the proposed method by approximating coefficients of shape profiles used in blocking noise reduction and inverse wavelet transform (WT) filters used in remainder noise reduction. The performance is evaluated for QCIF video sequences coded by H.263 TMN5 with quantization parameter (QP) in the range of 5-25 and is compared to that of the MPEG-4 verification model (VM) post-filter. Experimental results show that the proposed method yields not only PSNR improvement of maximum 0.5 dB over the VM post-filter but also subjective quality nearly free of the blocking artifact and edge blur.

In this paper, we analyze wavelet-based coding in a rate-distortion (R-D) sense by using Laplacian and Markov models and verify the results with the performance of the typical embedded coders, EZW and SPIHT, and the non-embedded coder implemented here. Laplacian represents the probability density function (pdf) of wavelet coefficients and Markov statistical dependency within and among subbands. The models allow us to easily understand the behavior of a thresholding and quantization part and a lossless coding part and associate the embedded coders with the nonembedded coder, which is the point the paper approaches. The analytic results are shown to coincide well with the actual coding results.

We present an efficient algorithm using a region-based texture feature for the extraction of texture regions. The key idea of this algorithm is to use the variations of local correlation coefficients (LCCs) according to different orientations to classify texture and shade regions. Experimental results show that the proposed feature suitably extracts the regions that appear visually as texture regions.