Jan P.H. Cornelis

Complete-to-overcomplete discrete wavelet transforms: theory and applications

A new transform is proposed that derives the overcomplete discrete wavelet transform (ODWT) subbands from the critically sampled DWT subbands (complete representation). This complete-to-overcomplete DWT (CODWT) has certain advantages in comparison to the conventional approach that performs the inverse DWT to reconstruct the input signal, followed by the a/spl grave/-trous or the lowband shift algorithm. Specifically, the computation of the input signal is not required. As a result, the minimum number of downsampling operations is performed and the use of upsampling is avoided. The proposed CODWT computes the ODWT subbands by using a set of prediction-filter matrices and filtering-and-downsampling operators applied to the DWT. This formulation demonstrates a clear separation between the single-rate and multirate components of the transform. This can be especially significant when the CODWT is used in resource-constrained environments, such as resolution-scalable image and video codecs. To illustrate the applicability of the proposed transform in these emerging applications, a new scheme for the transform-calculation is proposed, and existing coding techniques that benefit from its usage are surveyed. The analysis of the proposed CODWT in terms of arithmetic complexity and delay reveals significant gains as compared with the conventional approach.

Complete-to-overcomplete discrete wavelet transforms for scalable video coding with MCTF

Techniques for full scalability with motion-compensated temporal filtering (MCTF) in the wavelet-domain (in-band) are presented in this paper. The application of MCTF in the wavelet domain is performed after the production of the overcomplete discrete wavelet transform from the critically-sampled decomposition, a process that occurs at both the encoder and decoder side. This process, which is a complete-to-overcomplete discrete wavelet transform, is critical for the efficiency of the system with respect to scalability, coding performance and complexity. We analyze these aspects of the system and set the necessary constraints for drift-free video coding with in-band MCTF. As a result, the proposed architecture permits the independent operation of MCTF within different resolution levels or even different subbands of the transform and allows the successive refinement of the video information in resolution, frame-rate and quality.

New results on clutter reduction and parameter estimation for land mine detection using GPR

One of the main problems with the interpretation of GPR data is the strong ground reflection, obscuring signals arriving from just underneath the surface. The strength of this reflection can be reduced by deconvolution. This technique is especially useful when GPR is used to detect buried landmines. Parametric and non parametric time variant estimators are used for clutter characterization. Wavelet decomposition/reconstruction for noise removal is then applied. The application of the proposed signal processing technique to GPR data yields a substantial enhancement of the target reflection as well as a good estimation of physical parameters such as propagation velocity and target position.

Compression of volumetric medical data based on cube splitting

The increasing use of three-dimensional imaging modalities triggers the need for efficient techniques to transport and store the related volumetric data. Desired properties like quality and resolution scalability, region-of-interest coding, lossy-to-lossless coding and excellent rate-distortion characteristics for as well low as high bit-rates are inherently supported by wavelet-based compression tools. In this paper a new 3D wavelet-based compression engine is proposed and compared against a classical 3D JPEG-based coder and a state-of-the-art 3D SPIHT coder for different medical imaging modalities. Furthermore, we evaluate the performance of a selected set of lossless integer lifting kernels. We demonstrate that the performance of the proposed coder is superior for lossless coding, and competitive with 3D SPIHT at lower bit-rates.

Scalable motion vector coding

Modern video coding applications require transmission of video data over variable-bandwidth channels to a variety of terminals with different screen resolutions and available computational power. Scalable video coding is needed to optimally support these applications. Recently proposed wavelet-based video codecs employing spatial domain motion compensated temporal filtering (SDMCTF) provide quality, resolution and frame-rate scalability while delivering compression performance comparable to that of the state-of-the-art non-scalable H.264-codec. These codecs require scalable coding of the motion vectors in order to support a large range of bit-rates with optimal compression efficiency. Scalable motion vector coding algorithms based on the integer wavelet transform followed by embedded coding of the wavelet coefficients were recently proposed. In this paper, a new and fundamentally different scalable motion vector codec (MVC) using median-based motion vector prediction is proposed. Extensive experimental results demonstrate that the proposed MVC systematically outperforms the wavelet-based state-of-the-art solutions. To be able to take advantage of the proposed scalable MVC, a rate allocation mechanism capable of optimally dividing the available rate among texture and motion information is required. Two rate allocation strategies are proposed and compared. The proposed MVC and rate allocation schemes are incorporated into an SDMCTF-based video codec and the benefits of scalable motion vector coding are experimentally demonstrated.

Parallelization of the 2D fast wavelet transform with a space-filling curve image scan

The classical raster (i.e. row by row) image scan does not match the data processing flow, internal to the pyramid structure obtained by the 2D fast wavelet transform of a 2N X 2N image with a (2(gamma ) + 1) X (2(gamma ) + 1) mother wavelet, therefore introducing large latencies, important memory requirements, and irregular processor activities in parallelized implementations. A new algorithm is proposed in which all image data are scanned following a fractal, space-filling curve, which, compared to the raster image scan, offers the following advantages: i) it reduces the calculation memory with almost a factor 2, while maintaining a simple address calculation scheme, ii) the latency in the first N-(gamma) -3 levels of the pyramid, which contain a high percentage of the pyramid data, is minimized, leading to improved block-oriented post-processing capabilities (e.g. vector quantization for image compression), iii) the calculations are spread out more uniformly over one frame slot, and iv) the process is naturally subdivided into similar subproblems, increasing the granularity of the algorithm, without introducing severe communication bottle-necks for parallel architectures.

A wavelet-tree image coding system with efficient memory utilization

This paper describes an efficient implementation of an image coding system based on the independent wavelet-tree coding concept. The system consists of a transform and a (de)coding engine that operate in a pipelined fashion. The main focus of this paper is on the encoding part since, due to the system architecture, the decoder has identical memory utilization. Experimental results prove that the proposed system achieves comparable coding performance to the state-of-the-art, while it localizes the memory accesses to small memory modules and uses minimal computational resources.

Error protection and concealment of motion vectors in MCTF-based video coding

Error protection and concealment of motion vectors are of prime concern when video is transmitted over variable-bandwidth error-prone channels, such as wireless channels. In this paper, we investigate the influence of corrupted motion vectors in video coding based on motion-compensated temporal filtering, and develop various error protection and concealment mechanisms for this class of codecs. The experimental results show that our proposed motion vector coding technique significantly increases the robustness against transmission errors and generates performance gains of up to 7 dB compared with the original coding technique at the cost of less than 4% in terms of rate. It is also shown that our proposed spatial error-concealment mechanism leads to additional performance gains of up to 4 dB.

JPEG2000 Part 10: volumetric imaging

Recently, the JPEG2000 committee (ISO/IEC JTC1/SC29/WG1) decided to start up a new standardization activity to support the encoding of volumetric and floating-point data sets: Part 10 - Coding Volumetric and Floating-point Data (JP3D). This future standard will support functionalities like resolution and quality scalability and region-of-interest coding, while exploiting the entropy in the additional third dimension to improve the rate-distortion performance. In this paper, we give an overview of the markets and application areas targeted by JP3D, the imposed requirements and the considered algorithms with a specific focus on the realization of the region-of-interest functionality.

MAXAD distortion minimization for wavelet compression of remote sensing data

In the context of compression of high resolution multi-spectral satellite image data consisting of radiances and top-of-the-atmosphere fluxes, it is vital that image calibration characteristics (luminance, radiance) must be preserved within certain limits in lossy image compression. Though existing compression schemes (SPIHT, JPEG2000, SQP) give good results as far as minimization of the global PSNR error is concerned, they fail to guarantee a maximum local error. With respect to this, we introduce a new image compression scheme, which guarantees a MAXAD distortion, defined as the maximum absolute difference between original pixel values and reconstructed pixel values. In terms of defining the Lagrangian optimization problem, this reflects in minimization of the rate given the MAXAD distortion. Our approach thus uses the l-infinite distortion measure, which is applied to the lifting scheme implementation of the 9-7 floating point Cohen-Daubechies-Feauveau (CDF) filter. Scalar quantizers, optimal in the D-R sense, are derived for every subband, by solving a global optimization problem that guarantees a user-defined MAXAD. The optimization problem has been defined and solved for the case of the 9-7 filter, and we show that our approach is valid and may be applied to any finite wavelet filters synthesized via lifting. The experimental assessment of our codec shows that our technique provides excellent results in applications such as those for remote sensing, in which reconstruction of image calibration characteristics within a tolerable local error (MAXAD) is perceived as being of crucial importance compared to obtaining an acceptable global error (PSNR), as is the case of existing quantizer design techniques.