Search

This paper describes a new multi-resolution approach for the blind source separation of convolutive image mixtures in transform domain. The proposed method uses an Adaptive Quincunx Lifting Scheme based on wavelet decomposition and a Complex ICA unmixing algorithm. It proceeds in three steps: first, the mixed signals are decomposed by an adaptive lifting scheme. Then, the unmixing algorithm is applied to the more relevant component. The unmixed signals are, thereafter, reconstructed using an inverse transform. Experiments carried out on images from various origins showed that the proposed method yields better results than many widely used blind source separation algorithms.

This paper proposes an image enhancement method which is based on space-adaptive lifting scheme. The space-adaptive lifting scheme is designed with update-first strategy and the spatial domain adaptive Wiener filter is used for subband image enhancement. The space-adaptive lifting scheme is superior to the scale-adaptive one in that, the prediction stage is adapted to the signal structure pixel-by-pixel (not scale-by-scale) which results in a better signal representation and better image enhancement. A novel adaptive prediction method is proposed in the 2-D lifting scheme. The method adaptively chooses the best prediction filter (predictor), among a set of predictors, based on minimizing the prediction error (or detail coefficient). The experimental results and objective quality test results prove the accomplishment of the proposed image enhancement method.

This paper presents an adaptive lifting scheme, which performs integer-to-integer wavelet transform, for lossless image compression. We optimize the coefficients of the predict filter in the lifting scheme to minimize the entropy of the detail coefficients. The optimized coefficients depend on the auto-correlation structure of the image. The presented lifting scheme adapts to both horizontal and vertical directions of the image. We obtain experimental results using different types of images. These results show that the presented adaptive transform is a competitive wavelet transform for lossless image compression.

This paper investigates an adaptive lifting scheme for wavelet-based image compression, which performs update before prediction, and then we proposes an improved adaptive update and prediction scheme to better exploit the correlation among neighboring pixels. Simulation results show that the proposed scheme generates in the transformed domain lower coefficients' entropy, lower high-frequency coefficients' energy and more zero values, which make it useful for image compression applications.

International audience; Many existing works related to lossy-to-lossless image compression are based on the lifting concept. However, it has been observed that the separable lifting scheme structure presents some limitations because of the separable processing performed along the image lines and columns. In this paper, we propose to use a 2D non separable lifting scheme decomposition that enables progressive reconstruction and exact decoding of images. More precisely, we focus on the optimization of all the involved decomposition operators. In this respect, we design the prediction filters by minimizing the variance of the detail signals. Concerning the update filters, we propose a new optimization criterion which aims at reducing the inherent aliasing artefacts. Simulations carried out on still and stereo images show the benefits which can be drawn from the proposed optimization of the lifting operators.

In this paper, an improved adaptive update and adaptive prediction algorithm is put forward. The method makes full use of the correlation of neighboring pixels, which the approximation band and the detail bands obtained by a polyphase decomposition of an original image perform the proposed adaptive update step one by one, then the detail bands perform the adaptive prediction algorithm. The advantage of this method is its discrimination ability to use gradient operator for deciding the smooth areas and the discontinued areas, and choosing the different prediction filter according to the local characteristics. Simulation results show that the proposed lifting scheme algorithm generates lower entropy of high-frequency coefficients, lower energy of high-frequency coefficients and more zero values in wavelet transform domain, which makes it useful for digital image compression.

In the subband coding of image, direction of image features has already emphasized to want a function. In this paper, we propose a adaptive lifting transform (ALT) that adapts the direction filter to image based on the lifting scheme. In addition, We design a lossless and loss compression parallel system, which realized the energy of high frequency subbands as few as possible. Experimental results report outstanding to the conventional LWT for typical test images.

In this paper, a new adaptive lifting wavelet transform scheme is proposed. The main idea of the scheme is that the choice of predicting operator is triggered by local gradient information of the signal. Such an adaptive scheme can adapt itself well to the analyzed signal, and both the smooth properties and the sharp transition characteristics of the signal can be reserved perfectly by using this adaptive lifting wavelet transform. Simulation examples illustrate that the new scheme is effective.

The lifting scheme is well known to be an efficient tool for constructing second generation wavelets and is often used to design a class of bi-orthogonal wavelet filter banks. This paper presents the implementation of a class of Infinite Impulse Response (IIR) orthogonal wavelet filter banks by using the adaptive lifting scheme with two lifting steps. A class of II R orthogonal wavelet filter banks can be realized by using all pass filters in the lifting steps. The adaptive lifting scheme is based on the statistical characteristics of the image, the order of the prediction and updation filters are chosen by considering the statistical energy of the image. The designed IIR orthogonal wavelet filter banks have approximately linear phase responses. Finally, the proposed IIR orthogonal wavelet filter banks are applied to the image compression and the coding performance of the proposed IIR filter banks are analysed.

Because of the better temporal and frequency properties, the wavelet linear approximation can approximate smooth signals very efficiently with the constant size filter. However, standard wavelet linear approximation techniques cannot achieve similar results for signal which are not smooth. Based on the adaptive lifting scheme with perfect reconstruction presented by G. Piella and the author’s previous research, this paper proposed the improved double adaptive wavelet transform which can achieve the better linear approximation for the piecewise continuous signals.

De-noising by the traditional wavelet transform, the result is affected by the choosing of wavelet base. Because the wavelet base is fixed in the traditional wavelet transform, either the smoothness or singularity of the signal can’t be fitted quite well. To overcome the limitation, a new self-adaptive lifting scheme based on modulus maximum analysis is presented. Modulus maximum sequence of the large scale wavelet coefficients can locate the point of the signal with big singularity precisely. According to the position of the point with big singularity, proper neighborhood is fixed, and prediction operator can be chosen self-adaptively. In this way, the prediction operator is fitted to the local feature of the signal. The simulation and engineering application showed that the proposed method could overcome the de-noising disadvantage of traditional wavelet transform. It not only can filter noise from original signal effectively but also can hold local characteristics of original signal in the de-noised signals?

This paper presents an adaptive lifting scheme, which performs integer-to-integer wavelet transform, for lossless image compression. We optimize the coefficients of the predict filter in the lifting scheme to minimize the predictor's error variance. The optimized coefficients depend on the autocorrelation structure of the image. The presented lifting scheme adapts not only to every component of the color image, but also to its horizontal and vertical directions. We implement this lifting scheme on the fixed-point TMS320C6416 DSK evaluation board. We obtain experimental results using different types of images, as well as using images captured by camera in a real-time application. These results show that the presented method is competitive to few well-known methods for lossless image compression.

In this paper, we propose a new adaptive lifting scheme that not only locally adapts the filtering directions to the orientations of image features, but also adapts the lifting filters to the statistic properties of image signal. The proposed approach refines previous adaptive directional lifting-based wavelet transform (ADL) by combining directional lifting and adaptive lifting filters to form a unified framework. The image signal is first segmented into regions of textures and edges with close directional features. The lifting filters are then effectively designed. The prediction step is designed to minimize the prediction error of the image signal, and the update step is designed to minimize the reconstruction error. Significant improvements on objective and subjective quality over conventional 2-D wavelet transform and previous ADL transform are achieved.