Catálogo de publicaciones - libros

The newest video coding standard H.264/AVC provides multiple reference frames motion estimation in the spatial region, and the optimal frame is selected by RDO (Rate Distortion Optimization) with high coding complexity. However, the coding efficiency only depends on the attribute of sequences, not on the number of reference frames. In this paper, statistical characteristics of the best reference frame with variable block size are studied, and a fast algorithm that takes into account the correlation is proposed. The reference frame of block mode may be chosen based on the computing result of the above block mode. Experimental results show that with similar Distortion performance, the algorithm can efficiently reduce the computational complexity by 19% averagely.

The conventional side-view and rear-view mirrors are not enough for driver’s safety in an automobile. A driver may not be able to recognize the vehicle in a blind spot. In this paper, we propose an automotive detector algorithm using biologically motivated selective attention model for a blind spot monitor. This method decides a region of interest (ROI) which includes the blind spot from the successive image frames obtained by side-view cameras. It can detect the dangerous situations in the ROI using novelty points from the biologically motivated selective attention model, and alerts the driver whether there is dangerous object for changing the lane in driving. The proposed algorithm is based on deciding the ROI using difference from intensity histogram of a Gaussian smoothed image and finding the novelty points from the biologically motivated selective attention model. From variations of those novelty points, we determine whether a vehicle is approaching or not.

Though wavelet transform based methods have recently raised increasing interests in texture analysis due to their good space and frequency localization, many issues related to the choice of the wavelet basis and texture feature remain unresolved. In this paper, we evaluate the performance of seven wavelet energy signatures and eight wavelet basis for texture discrimination. Experimental results on 111 Brodatz textures show that the feature extracted from high and middle frequency channels is more suitable for texture analysis and the choice of wavelet basis has some influence on texture discrimination.

On the basis of analyzing the performances of popular image fusion methods, a new remote sensing image fusion method based on principal component analysis (PCA), high pass filter (HPF) and undecimated discrete wavelet transform (UDWT) is proposed. Some measure parameters are suggested to evaluate the fusion method. Experiments have been performed with the SPOT panchromatic image and the TM multi-spectral image. Both subjectively qualitative analysis and objectively quantitative evaluation verify the performance of the new method. With the same wavelet transform level, the fusion image using the proposed method preserves more sophisticated spatial details and distorts less spectral information in comparison with the fusion image using the traditional discrete wavelet transform (DWT) method.

Recent researches have been focusing on fusion of audio and visual features for reliable speech recognition in noisy environments. In this paper, we propose a neural network based model of robust speech recognition by integrating audio, visual, and contextual information. Bimodal Neural Network (BMNN) is a multi-layer perceptron of 4 layers, which combines audio and visual features of speech to compensate loss of audio information caused by noise.  In order to improve the accuracy of speech recognition in noisy environments, we also propose a post-processing based on contextual information which are sequential patterns of words spoken by a user. Our experimental results show that our model outperforms any single mode models. Particularly, when we use the contextual information, we can obtain over 90% recognition accuracy even in noisy environments, which is a significant improvement compared with the state of art in speech recognition.

An analysis based on wavelet modulation scales feature extraction is proposed. Considering human auditory perception and varieties of disturbances, instead of the frequency differences, wavelet modulation scales are adopted to reflect the dynamic features of speech in ASR. Experiments for the Chinese digit-string recognition show extracting the wavelet modulation scales as the dynamic features have good performance both in additional noises and convolutional noises environment.

As multimedia and group-oriented computing becomes increasingly popular for the users of mobile ad hoc networks (MANET). Due to the dynamic nature of the network topology and restricted resources, quality of service (QoS) and multicast routing in MANET is a challenging task. It attracts the interests of many people. In this paper, we present a fuzzy controllers based QoS routing algorithm with a multiclass scheme (FQRA) in MANET. The performance of this scheduler is studied using NS2 and evaluated in terms of quantitative met-rics such as path success ratio, average end-to-end delay and throughput. Simu-lation shows that the approach is efficient, promising and applicable in MANET.

Many high resolution DOA estimation algorithms like MUSIC and ESPRIT estimation are based on the sub-space concept and require the eigen-decomposition of the input correlation matrix. As quantities of computation of eigen-decomposition, it is unsuitable for real time processing. An algorithm for noise subspace estimation based on minor component analysis is proposed. These algorithms are based on anti-Hebbian learning neural network and contain only relatively simple operations, which are stable, convergent, and have self-organizing properties. Finally a method of real-time parallel processing is proposed, and data processing can be finished at end time of sampling. Simulations show that the proposed algorithm has an analogy performance with the MUSIC algorithm.

To extract source signals with certain temporal structures, such as periodicity, we propose a two-stage extraction algorithm. Its first stage uses the autocorrelation property of the desired source signal, and the second stage exploits the independence assumption. The algorithm is suitable to extract periodic or quasi-periodic source signals, without requiring that they have distinct periods. It outperforms many existing algorithms in many aspects, confirmed by simulations. Finally, we use the proposed algorithm to extract the components of visual event-related potentials evoked by three geometrical figure stimuli, and the classification accuracy based on the extracted components achieves 93.2%.

This paper examines the relation between the degree of experimentally induced focal ischemia in the left-brain of 24 experimental rats and Higher Order Statistics (HOS) such as the bispectrum and the bicoherence index of scalp EEG recorded at the time of the ischemic event. The aim is to propose the assessment of HOS in non-invasive scalp EEG to facilitate identification and even classification of focal ischemic events in terms of the degree of tissue damage. The latter is achieved by a supervised, multilayer, feed-forward Artificial Neural Network (ANN). The ANN utilizes a back propagation algorithm to classify ischemic states of the brain. The target values used during the training session of the network are the degree of ischemic tissue damage (graded as serious, middle and slight) as assessed by histological and immunhistochemical methods in the brain slice of the experimental animals. The results show that the ANN can correctly identify and classify ischemic events with high precision 91.67% based on HOS measures of scalp EEG obtained during ischemia. These findings may potentially be of great scientific merit, especially due to their possibly very important medical implications: a potential non-invasive method that reliably identifies the presence and the degree of ischemia at the time of its occurrence.