Search

Nematode contagion is a core impediment to the profitable dairy production for livestock farms. Gastrointestinal (GI) parasitism causes weight loss and low milk production, along with high mortality in sheep and goat. The current intensive anthelmintics therapies to control parasitic burden resulted in the appearance of drug resistant parasitic strains. Due to high demand of unadulterated animal products free from drugs motivated alternative strategies for improvement; including breeding plans for parasite control in ruminants. The increase of protective resistance to nematode infections, host genome show particular expression that is frequently confused by mechanisms concurrently essential to control multiple nematodes species as well as protozoan ecto-parasites, viral and microbial pathogens. The involved molecular mechanisms under these developments correspond to crucial steps toward improvement of efficient new parasite control strategies. Knowledge of various immunity methods of host and regulation of development of parasite, physiology, and virulence is able to identify the objective of parasite control. This review recapitulate current evolution and restrictions of optimistic regulatory biological pathways and genetic networks that concern with susceptibility and host resistance to infection of GI nematode in ruminants.

In recent years, the scale of the wireless network of communication operators has been expanding. Various WLAN (Wireless Local Area Networks) hotspots are increasing and the application scenarios are becoming more diverse. The traditional WLAN network testing has limitations of cost, application scenarios, and sampling points. With the rapid popularization of smart phones, their functions are becoming increasingly powerful and intelligent. As a simple and portable functional mobile device, the development of the smart phone provides a better platform for portable wireless detection. In this paper, a software design and implementation method of WLAN sensing application based on an Android intelligent terminal is proposed. The developed sensing APP client implements the WLAN hotspot network quality testing and technology validation, providing a low-cost and efficient way to realize the development of the nationwide wireless network perception.

We propose a tracking method using both overall information of object and local representation.The object is divided into m overlapped patches to improve tracking performance in background clutter and partial occlusions.We take the prior information into account for avoiding ambiguity in discriminative model.The dictionary is updated by Metropolis-Hastings to capture the appearance changes.We achieve comparable performances with other methods on challenging sequences. In this paper, we propose a novel visual tracking algorithm using the collaboration of generative and discriminative trackers under the particle filter framework. Each particle denotes a single task, and we encode all the tasks simultaneously in a structured multi-task learning manner. Then, we implement generative and discriminative trackers, respectively. The discriminative tracker considers the overall information of object to represent the object appearance; while the generative tracker takes the local information of object into account for handling partial occlusions. Therefore, two models are complementary during the tracking. Furthermore, we design an effective dictionary updating mechanism. The dictionary is composed of fixed and variational parts. The variational parts are progressively updated using Metropolis-Hastings strategy. Experiments on different challenging video sequences demonstrate that the proposed tracker performs favorably against several state-of-the-art trackers.

In this paper, we propose a two-stage sparse decomposition-based method for single-channel speech separation in time domain. First, we propose a Dictionary-updated orthogonal matching pursuit (DUOMP) algorithm which is used in both separation stages. In the proposed DUOMP algorithm, all atoms of each source-specific dictionary are updated by subtracting off the current approximation of each source to the original atoms. It is proved that the DUOMP algorithm can limit the separated sources within a region where they are uncorrelated in statistical sense more quickly. Then, we propose an adaptive dictionary generation method followed by a frame labeling method to perform a second-stage separation on the mixed frames having certain temporal structure. Experiments show that the proposed method outperforms a separation method using sparse non-negative matrix factorization (SNMF), a separation method using OMP and a source-filter-based method using pitch information in overall. Additionally, what affects the performance of the proposed method is also shown.

Recognizing human interactions is a more challenging task than recognizing single person activities and has attracted much attention of the computer vision community. This paper proposes an innovative and effective way to recognize human interactions, which incorporates the advantages of both global motion context (MC) feature and spatio-temporal (S-T) correlation of local spatio-temporal interest point feature. The MC feature is used to train a random forest where genetic algorithm (GA) is applied to the training phase to achieve a good compromise between reliability and efficiency. Besides, we propose S-T correlation-based match, where MC's structure and Needleman---Wunsch algorithm are used to calculate the spatial and temporal correlation score of two videos, respectively. Experiments on the UT-Interaction dataset show that our approaches outperform other prevalent machine learning methods, and that the combination of GA search-based random forest and S-T correlation achieves the state-of-the-art performance.

Nowadays, local features are very popular in vision-based human action recognition, especially in "wild" or unconstrained videos. This paper proposes a novel framework that combines Fast HOG3D and self-organization feature map (SOM) network for action recognition from unconstrained videos, bypassing the demanding preprocessing such as human detection, tracking or contour extraction. The contributions of our work not only lie in creating a more compact and computational effective local feature descriptor than original HOG3D, but also lie in first successfully applying SOM to realistic action recognition task and studying its training parameters' influence. We mainly test our approach on the UCF-YouTube dataset with 11 realistic sport actions, achieving promising results that outperform local feature-based support vector machine and are comparable with bag-of-words. Experiments are also carried out on KTH and UT-Interaction datasets for comparison. Results on all the three datasets confirm that our work has comparable, if not better, performance comparing with state-of-the-art.

We propose a novel approach for visual tracking based on a particle swarm optimization (PSO) framework using SIFT feature points correspondence and multiple fragments in a candidate target region to cope with the problems of partial occlusions, illumination changes, and large motion changes of the tracked target. Firstly, optimal search in the successive frame tracking process is performed by the PSO algorithm, which guides all particles towards the global optima state based on a fitness function. Then, the SIFT feature information is integrated into the iterative results of PSO to acquire a more accurate tracking state. Secondly, we present an effective appearance model updating criterion, which evaluates which fragments in appearance model need updating at each frame. However, the fragments with occluded parts or low quality measure values are not updated. The method for updating appearance model is introduced to improve the tracking performance. Compared with state-of-the-art algorithms, the proposed method can still stably track the target during the course of long-term partial occlusions using superior fragments of tracked target. The experiment results demonstrate the effectiveness of our algorithm in complex environments where the target object undergoes partial occlusions and large changes in pose and illumination.

In this paper, we describe an adaptive object detection system based on boosted weak classifiers. We formulate the learning of object detection as to find the representative sharing features over the examples of object. Objects with low intra-class variation imply that more generic features for detection can be selected. In contrast, more specific features for only subset of examples should be encouraged to gain an elegant representation for the object with high intra-class variation. In this spirit, we implement an implicit partition over positive examples to obtain a data-driven clustering. Based on the implicit partition, we encourage an intra-class and inter-class feature sharing between sub-categories and build an adaptive hierarchical cascade of weak classifiers. Experimental results prove that the intra-class sharing makes an adaptive trade-off between performance and efficiency on various object detection tasks. The inter-class sharing allows objects to borrow semantic information from related well labeled objects. We hope that the proposed implicit feature sharing over sub-categories can extend the application of traditional boosting methods.

In this paper, we propose a model adaptation algorithm based on maximum likelihood subband polynomial regression (MLSPR) for robust speech recognition. In this algorithm, the cepstral mean vectors of prior trained hidden Markov models (HMMs) are converted to the log-spectral domain by the inverse discrete cosine transform (DCT) and each log-spectral mean vector is divided into several subband vectors. The relationship between the training and testing subband vectors is approximated by a polynomial function. The polynomial coefficients are estimated from adaptation data using the expectation–maximization (EM) algorithm under the maximum likelihood (ML) criterion. The experimental results show that the proposed MLSPR algorithm is superior to both the maximum likelihood linear regression (MLLR) adaptation and maximum likelihood subband weighting (MLSW) approach. In the MLSPR adaptation, only a very small amount of adaptation data is required and therefore it is more useful for fast model adaptation.

Sound source localization plays a crucial role in many microphone arrays application, ranging from speech enhancement to human–computer interface in a reverberant noisy environment. The steered response power (SRP) using the phase transform (SRP-PHAT) method is one of the most popular modern localization algorithms. The SRP-based source localizers have been proved robust, however, the methods may fail to locate the sound source in adverse noise and reverberation conditions, especially when the direct paths to the microphones are unavailable. This paper proposes a localization algorithm based on discrimination of cross-correlation functions. The cross-correlation functions are calculated by the generalized cross-correlation phase transform (GCC-PHAT) method. Using cross-correlation functions, sound source location is estimated by one of the two classifiers: Naive-Bayes classifier and Euclidean distance classifier. Simulation results have demonstrated that the proposed algorithms provide higher localization accuracy than the SRP-PHAT algorithm in reverberant noisy environment.

Topic models have been wildly applied in the field of computer vision, through which superior performance was yielded in various recognizing tasks. Among them, probabilistic latent semantic analysis model has earned much attention due to its simplicity and effect. But the affection of encoding and normalization methods on topic models has been ignored during the period. This paper explores the impact of encoding methods combined with different normalization on probabilistic latent semantic analysis model in the context of action classification in videos. Detailed experiments are conducted on KTH and UT-interaction datasets. The results show that an appropriate combination of encoding and normalization methods could significantly improve the performance of probabilistic latent semantic analysis model. The recognition accuracy reachs 96.44% and 93.33% on UT-interaction set1 and set2 respectively, which outperforms the state-of-the-art. Especially, we obtain 94.24% on UT-interaction set1 using sparse STIPs.

Sound source localization is essential in many microphone arrays application, ranging from teleconferencing systems to artificial perception in a reverberant noisy environment. The steered response power (SRP) using the phase transform (SRP-PHAT) source localization algorithm has been proved robust, however, the performance of the SRP-PHAT algorithm degrades in highly reverberant noisy environment. Though the SRP-based maximum likelihood localizers are more robust than SRP-PHAT, they have the drawback of requiring noise variance to be estimated in a silent room. This paper presents an improved SRP-PHAT algorithm based on principal eigenvector. Sound source location is estimated from the principal eigenvector computed from the frequency-domain correlation matrix. Using both simulated and real data, we show that the proposed algorithm achieves higher source localization accuracy compared to the SRP-PHAT algorithm.

This paper proposes a novel robust speech recognition technique using improved vector Taylor series (VTS) algorithm for embedded systems. It uses a hidden Markov model (HMM) to replace the Gaussian mixture model (GMM) for estimating the clean speech feature, and gives the closed-form solutions of the noise parameters including the mean and variance at each expectation-maximization (EM) iteration. The experimental results show that the proposed algorithm makes a good balance between the computational complexity and recognition accuracy, and thus is more useful for embedded systems.

In microphone arrays application, it is a difficult task to accurately and fast localize sound source in a noisy, reverberant environment. In order to solve this problem, many approaches have been presented. Among them, the steered response power-phase transform weighted (SRP–PHAT) source localization algorithm has been proved robust. However, SRP–PHAT requires high computation cost for searching a large location space. To overcome this shortcoming, an improved SRP–PHAT will be presented that reduces a two-dimension searching space into a couple of one-dimension ones by using an orthogonal linear array. In this method, the parameters of direction of arrival (DOA) are separated. The main computation can be carried out independently in two one-dimension spaces, thus the computational load will be greatly cut down. Simulations show that there is no loss in accuracy in the proposed method.

If a somewhat fast moving object exists in a complicated tracking environment, snake’s nodes may fall into the inaccurate local minima. We propose a mean shift snake algorithm to solve this problem. However, if the object goes beyond the limits of mean shift snake module operation in successive sequences, mean shift snake’s nodes may also fall into the local minima in their moving to the new object position. This paper presents a motion compensation strategy by using particle filter; therefore a new Particle Filter Mean Shift Snake (PFMSS) algorithm is proposed which combines particle filter with mean shift snake to fulfill the estimation of the fast moving object contour. Firstly, the fast moving object is tracked by particle filter to create a coarse position which is used to initialize the mean shift algorithm. Secondly, the whole relevant motion information is used to compensate the snake’s node positions. Finally, snake algorithm is used to extract the exact object contour and the useful information of the object is fed back. Some real world sequences are tested and the results show that the novel tracking method have a good performance with high accuracy in solving the fast moving problems in cluttered background.

The synthesis of individual virtual auditory space (VAS) is an important and challenging task in virtual reality. One of the key factors for individual VAS is to obtain a set of individual head related transfer functions (HRTFs). A customization method based on back-propagation (BP) artificial neural network (ANN) is proposed to obtain an individual HRTF without complex measurement. The inputs of the neural network are the anthropometric parameters chosen by correlation analysis and the outputs are the characteristic parameters of HRTFs together with the interaural time difference (ITD). Objective simulation experiments and subjective sound localization experiments are implemented to evaluate the performance of the proposed method. Experiments show that the estimated non-individual HRTF has small mean square error, and has similar perception effect to the corresponding one obtained from the database. Furthermore, the localization accuracy of personalized HRTF is increased compared to the non-individual HRTF.

Super-resolution reconstruction (SRR) produces a high-resolution image from multiple low-resolution images. Many image SRR algorithms assume that the blurring process, i.e., point spread function (PSF) of the imaging system is known in advance. However, the blurring process is not known or is known only to within a set of parameters in many practical applications. In this letter, we propose an approach for solving the joint blur identification and image SRR based on the principle similar to the variable projection method. The approach can avoid some shortcomings of cyclic coordinate descent optimization procedure. We also propose an efficient implementation based on Lanczos algorithm and Gauss quadrature theory. Experimental results are presented to demonstrate the effectiveness of our method.

A novel text independent speaker identification system is proposed. In the proposed system, the 12-order perceptual linear predictive cepstrum and their delta coefficients in the span of five frames are extracted from the segmented speech based on the method of pitch synchronous analysis. The Fisher ratios of the original coefficients then be calculated, and the coefficients whose Fisher ratios are bigger are selected to form the 13-dimensional feature vectors of speaker. The Gaussian mixture model is used to model the speakers. The experimental results show that the identification accuracy of the proposed system is obviously better than that of the systems based on other conventional coefficients like the linear predictive cepstral coefficients and the Mel-frequency cepstral coefficients.

Trajectories extracted by previous methods for human action recognition contain irrelevant changes, and the Orientation-Magnitude descriptors of their shapes lack the robustness to camera motion. To solve these problems, action recognition by tracking salient relative motion points is proposed in this paper. Firstly, motion boundary detector which suppresses the camera constant motion is utilized to extract motion features. After processing the detected boundaries by the adaptive threshold, the super-pixels that contain salient points are defined as relative motion regions. Then tracking the points within super-pixels is to generate trajectories. For the trajectory shape, the pre-defined orientation assignments with coarse-to-fine quantization levels are used to produce orientation statistics. Finally, the descriptors of oriented gradient, motion boundary, oriented statistic and their combination are adopted to represent action videos, respectively. On the benchmark KTH and UCF-sports action datasets, experimental results show that the extracted trajectories can describe the movement process of object. Compared with the conventional algorithms, our method with multiple kernel learning obtains good performance.

Human action recognition is at the core of computer vision, and has great application value in intelligent human-computer interactions. On the basis of Bag-of-Words (BoW), this work presents a Huffman coding and Implicit Action Model (IAM) combined framework for action recognition. Specifically, Huffman coding, which outperforms naive Bayesian method, is a robust estimation of visual words' conditional probabilities; whereas IAM captures the spatio-temporal relationships of local features and outperforms most other common machine learning methods. Spatio-Temporal Interest Points (STIPs) and Harris corners are employed as local features, and multichannel feature description is adopted to exploit the complementarity among different features. Experiments on UCF-YouTube and HOHA2 datasets systematically compare the performance of various feature channels and machine learning methods, demonstrating the effectiveness of the approaches proposed by this paper. Finally, multiple augment mechanisms such as feature fusion, hierarchical codebooks and sparse coding are integrated into the recognition system, achieving the best ever performance comparing with the state-of-the-art.

To a significant degree, multimedia applications derive their effectiveness from the use of color graphics, images, and videos. In these applications, human visual system (HVS) often gives the final evaluation of the processed results. In this paper, we first propose a novel color image enhancement method, which is named HVS Controlled Color Image Enhancement and Evaluation algorithm (HCCIEE algorithm). We then applied the HCCIEE to color image by considering natural image quality metrics. This HCCIEE algorithm is base on multiscale representation of pattern, luminance, and color processing in the HVS. Experiments illustrated that the HCCIEE algorithm can produce distinguished details without ringing or halo artifacts. (These two problems often occur in conventional multiscale enhancement techniques.) As a result, the experimental results appear as similar as possible to the viewers' perception of the actual scenes.

Recent research in transform-based image denoising has focused on the wavelet transform due to its superior performance over other transform. Performance is often measured solely in terms of PSNR and denoising algorithms are optimized for this quantitative metric. The performance in terms of subjective quality is typically not evaluated. Moreover, human visual system (HVS) is often not incorporated into denoising algorithm. This paper presents a new approach to color image denoising taking into consideration HVS model. The denoising process takes place in the wavelet transform domain. A Contrast Sensitivity Function (CSF) implementation is employed in the subband of wavelet domain based on an invariant single factor weighting and noise masking is adopted in succession. Significant improvement is reported in the experimental results in terms of perceptual error metrics and visual effect.

This paper introduces a novel algorithm to image enhancement that exploits the multi-scale wavelet and statistical characters of visual representation. Processing includes the global dynamic range (brightness) correction and local contrast adjustment, whose parameters are picked automatically by the information contained in the image itself. Experimental results show that the new algorithm outperforms other many existing image enhancement methods and is highly resilient to the effects of both the image-source variations. (C) 2004 Elsevier B.V. All rights reserved.

In this paper, we propose a new sparse decomposition based single-channel speech separation method using orthogonal matching pursuit (OMP). The separation is performed using source-individual dictionaries consisting of time-domain training frames as atoms. OMP is used to compute sparse coefficients to estimate sources. We report the separation results of our proposed method and compare them with a separation method based on sparse non-negative matrix factorization (SNMF) which is a classical sparse decomposition based separation method. Experiments show that our proposed method results in higher signal-to-noise ratio (SNR) and signal-to-interference ratio (SIR).

Handling appearance variations caused by the occlusion or abrupt motion is a challenging task for visual tracking. In this paper, we propose a novel tracking method that deals with the appearance changes based on sparse representation in a particle swarm optimization (PSO) framework. First, we divide each candidate state into multiple structural patches to cope with the partial occlusions of the object. Once the object is lost, we present an object’s recovery scheme by the scale invariant feature transforms (SIFT) correspondence between two frames to reacquire the rough object position. Then the tracking state is searched in the vicinage of the rough object position using the PSO iteration. In addition, an online dictionary updating mechanism is presented to capture the object appearance variations. The object information from the initial frame is never updated in the tracking, while other templates in the dictionary are progressively updated based on the coefficients of templates. Compared with several conventional trackers, the experimental results demonstrate that our approach is more robust in dealing with the occlusions and abrupt motion variations.

This paper proposes an innovative and effective hybrid way to recognize human interactions, which incorporates the advantages of both global feature (Motion Context, MC) and Spatio-Temporal (S-T) correlation of local Spatio-Temporal Interest Points (STIPs). The MC feature, which also derives from STIPs, is used to train a random forest where Genetic Algorithm (GA) is applied to the training phase to achieve a good compromise between reliability and efficiency. Besides, we design an effective and efficient S-T correlation based match to assist the MC feature, where MC's structure and a biological sequence matching algorithm are employed to calculate the spatial and temporal correlation score, respectively. Experiments on the UT-Interaction dataset show that our GA search based random forest and S-T correlation based match achieve better performance than some other prevalent machine leaning methods, and that a combination of those two methods outperforms most of the state-of-the-art works.

Sparse representation has been shown competitive performance on single object tracking. In this paper, we extend this technique to tracking multiple interactive objects and present a novel sparse tracker under the tracking-by-detection framework, with saliency detector for objects detection and sparse representation for objects association. Furthermore, we propose an online dictionary learning scheme to capture appearance variations of objects. To avoid using trivial templates, the dictionary contains not only objects templates, but also background information, resulting in more robust estimation. The experiments demonstrate that our approach achieves favorable performance over state-of-the-art algorithms.

This paper explores the performance of Neural Network (NN) for human action recognition and proposes a novel hierarchical and boosting-based action recognition system. Specifically, the main contributions of our work are three-fold: (1) A boosted NN based scheme is applied to the human action recognition task for the first time, during which we extend the standard binary AdaBoost algorithm to a multiclass version; (2) A novel hierarchical recognition framework with pre-decision and post-decision modules is proposed, which can significantly enhance the training efficiency as well as the frame-based recognition accuracy; (3) Numerous modified features (both motion and shape features) are utilized and combined in this paper. Experiments on the Weizmann dataset show promising results of our approach in comparison with other state-of-the-art methods.

Despite the success of many advanced tracking methods in this area, tracking targets with drastic variation of appearance such as deformation, view change and partial occlusion in video sequences is still a challenge in practical applications. In this letter, we take these serious tracking problems into account simultaneously, proposing a dynamic graph based model to track object and its deformable parts at multiple resolutions. The method introduces well learned structural object detection models into object tracking applications as prior knowledge to deal with deformation and view change. Meanwhile, it explicitly formulates partial occlusion by integrating spatial potentials and temporal potentials with an unparameterized occlusion handling mechanism in the dynamic conditional random field framework. Empirical results demonstrate that the method outperforms state-of-the-art trackers on different challenging video sequences., Comment: 4 pages, 5 figures, the manuscript has been submitted to IEEE Signal Processing Letters

In this paper, we address the problem of using boosting to detect different classes of objects with significant intra-class variation. Current approaches generally require prior knowledge such as semantic information to obtain an explicit partition over positive samples. This can require a lot of manually labeled samples and can limit the performance of classifier due to subjective error. We present a novel JointBoost based learning method which can learn the variant appearances of targets without any prior assumption. With an implicit partition, the specific features for fractional samples can contribute to classification as well as generic features for the whole training set. By encouraging an intra-class and inter-class feature sharing between implicit sub-categories, our data-driven learning approach avoids a local optimum in the candidate weak classifier space. Experimental results on two popular tasks demonstrate the considerable improvements brought by the new approach. We hope that implicit JointBoost will extend the application of traditional boosting methods.

The steered response power with phase transform weighted (SRP-PHAT) is a robust sound source localization method based on microphone array. Multiple source localization has been implemented using SRP-PHAT with agglomerative clustering (AC). In this paper, a novel method of multiple speech source localization based on sub-band SRP is proposed. In this method, speech signal is divided into several sub-bands, where sub-band SRP is computed, initial estimations are generated by searching the maximum in every sub-band SRP, and the final source locations are determined from the initial estimations using AC. The proposed method is tested with the real-world recordings under the condition that the number of active speakers is unknown. The results show that our method provides higher localization performance than that of the conventional SRP-PHAT with AC in the cases with up to 3 concurrent speakers.

Adaptive boosting (AdaBoost) learning method can improve the performance of a base classifier by mining feature information in depth. But it is computationally expensive, and the base classifier without a suitable accuracy will cause over fitting. In this paper an improved Adaboost algorithm using maximum a posteriori vector quantization model (VQMAP) for speaker identification is presented. A suitable VQMAP classifier matched the size of speaker identification problem is constructed first. Then it is boosted to a strong classifier by AdaBoost with early stopping method. Experiments show that the performance of the boosted VQMAP classifier is better than that of VQMAP, and is slightly lower than that of maximum a posteriori adapted Gaussian mixture model (GMMMAP), but with a faster recognition speed. In the case of limited data and predictable speaker number, it will reach or exceeded GMMMAP.

This paper addresses the problem of small scale target tracking. The divided-by-zero problem in the weight computation of mean shift algorithm and its associated tracking interrupt problem are presented. To tackle these problems, the Parzen window density estimation method is modified to interpolate the histogram of the target candidate. Then the Kullback-Leibler distance is employed as a new similarity measure between the target model and the target candidate. Its corresponding weight computation and new location expressions are derived. On the basis of these works, we propose a new small target tracking algorithm using mean shift framework. The tracking experiments for real world video sequences show that the proposed algorithm can track the target successively and accurately. It can successfully track very small targets with only 6×12 pixels.

The synthesis of individual virtual auditory system (VAS) is an important and state-of-the-art technology in virtual reality. One of the key factors for individual VAS is to obtain a set of proper head related transfer functions (HRTFs). A personalization method is presented in this paper. First, multiple linear regression analysis is applied to get the linear relationship between HRTFs and some anthropometric parameters; second, the magnitude response of the individual HRTF is estimated; finally the center frequencies of the two prominent notches in the estimated magnitude response are modified by using the center frequencies estimated from the subject's pinna parameters. The results show that the modified HRTF is more approximate to the measured one.

This paper addresses the problem of reconstructing a super- resolved image from a set of aliased, noisy, low-resolution (LR), and blurry images. The accurate knowledge of the sub-pixel registration parameters for each LR image is the key for this problem. However, the presence of aliasing is the main challenge in registering these images. In this paper, we analyze the method which combines the registration problem into the super-resolution reconstruction (SSR) and propose a novel approach to solve it. The proposed approach utilizes the principle similar to the variable projection method, which results in a better-conditioned problem and avoids some shortcomings of cyclic coordinate descent optimization procedure. It can be efficiently implemented by using Lanezos algorithm and Gauss quadrature theory. As a result, the proposed approach can deal with translation and rotation between the observed LR images. Experimental results demonstrate the effectiveness of our approach.

The synthesis of personalization virtual auditory space (VAS) is an important and challenging task in virtual reality. One of the key factors for personalization VAS is to obtain a set of proper head related transfer functions (HRTFs). In this paper, a customization method is presented by applying multiple linear regression analysis to HRTFs and some anthropometric parameters chosen by correlation analysis. Experiments show that the estimated HRTF has small mean square error, and has similar perception effect as the measured HRTF. Furthermore, the localization accuracy of personalized HRTF is increased compared to the non-individual HRTF.

Video object extraction has been a hot topic in the field of modern information processing. This paper proposes a novel video object watershed segmenting strategy, which combines the merits of alternative sequential filtering by reconstruction and adaptive threshold algorithm. This strategy can automatically determine the structural element size of the morphological filter and the threshold value of nonlinear transformation as well. The iteration of opening-closing by reconstruction with gradually expanding structural elements guarantees the reduction of the segmented regions. Besides, the iteration makes it easier to distinguish the gradient of edge pixels and that of the pixels inside the plateaus, which facilitates removing the small regions of local gradient extrema by adaptive threshold algorithm. Experiment results show that the proposed strategy brings about satisfactory spatial segmenting results.

A novel gender classification system has been proposed based on Gaussian Mixture Models, which apply the combined parameters of pitch and 10th order relative spectral perceptual linear predictive coefficients to model the characteristics of male and female speech. The performances of gender classification system have been evaluated on the conditions of clean speech, noisy speech and multi-language. The simulations show that the performance of the proposed gender classifier is excellent; it is very robust for noise and completely independent of languages; the classification accuracy is as high as above 98% for all clean speech and remains 95% for most noisy speech, even the SNR of speech is degraded to 0dB.

The mean shift algorithm is an efficient technique for object tracking. However, it has a shortcoming that it can't adjust scale with object during tracking process. There are presently no effective ways to solve this problem. The kernel bandwidth of mean shift tracker in one frame is generally steered by the object scale obtained in the previous frame, so it is very important for mean shift tracker to correctly describe the scale of the target in very frame. In accordance with the kernel-bandwidth effect on the mean shift tracker and the property of face, this paper introduces a new idea that uses direct least square ellipse fitting to adjust the facial scale. The experimental results demonstrate the efficiency of this algorithm. Its performance has been proven superior to the original mean shift tracking algorithm.

According to auditory characteristics of human's hearing system, a multi-band maximum a posteriori multi-transformation algorithm based on the discriminative combination is developed to improve the performance of speech recognition systems in noisy environment. The algorithm utilizes the difference between noise's spectrum and speech's spectrum, and the different effects of noise on recognition performance in different frequency bands. It compensates the effect of noise with a discriminative function and maximum a posteriori multi-transformation. Experimental results show that the proposed algorithm outperforms the maximum a posterior linear regression algorithm. The results also show that the utilization of effective band with information redundancy helps to improve the recognition performance.

The performance of speech recognition system will be significantly deteriorated because of the mismatches between training and testing conditions. This paper addresses the problem and proposes an algorithm to adapt the mean and covariance of HMM simultaneously within the minimum classification error linear regression (MCELR) framework. Rather than estimating the transformation parameters using maximum likelihood estimation (MLE) or maximum a posteriori, we proposed to use minimum classification error (MCE) as the estimation criterion. The proposed algorithm, called IMCELR (Improved MCELR), has been evaluated on a Chinese digit recognition tasks based on continuous density HMM. The experiments show that the proposed algorithm is more efficient than maximum likelihood linear regression with the same amount of adaptation data.

We propose a novel color image enhancement method, the human visual system controlled color image enhancement and evaluation (HCCIEE) algorithm, and apply it to color images by considering natural image quality metrics. This HCCIEE algorithm is based on a multiscale representation of pattern, luminance and color processing in the human visual system. Experiments illustrate that the HCCIEE algorithm can produce distinguishing details while avoiding artifacts, which often occur in conventional multiscale enhancement methods, as well as producing images that appear as similar as possible to the viewer's perception of actual scenes.

Recent research in transform-based image denoising has focused on the wavelet transform due to its superior performance over other transform. Performance is often measured solely in terms of PSNR and denoising algorithms are optimized for this quantitative metric. The performance in terms of subjective quality is typically not evaluated. Moreover, human visual system (HVS) is often not incorporated into denoising algorithm. This paper presents a new approach to color image denoising taking into consideration HVS model. The denoising process takes place in the wavelet transform domain. A Contrast Sensitivity Function (CSF) implementation is employed in the subband of wavelet domain based on an invariant single factor weighting and noise masking is adopted in succession. Significant improvement is reported in the experimental results in terms of perceptual error metrics and visual effect. (C) 2004 Elsevier B.V. All rights reserved.

The directional-dependent information in the head-related transfer function (HRTF) is important for the study of human sound localization system and the synthesis of virtual auditory signals. Its time-domain and frequency-domain characteristics have been widely studied by researchers. The purpose of this paper is to explore the ability of discrete wavelet transform to describe the time-scale characteristics of HRTFs. Both the time-domain characteristics and energy distribution of different frequency subbands were studied. Discrete wavelet analysis is found to be a new direction-dependence information showing the relation of the characteristics of the HRTFs to sound source directions.

A virtual acoustic space (VAS) employs the localization cues specified by the direction-dependent 'free-field to eardrum transfer function' (FETF) to synthesize sound-pressure waveforms present near the tympanum. The combination of a VAS and the earphone delivery of synthesized waveforms is useful to study parametrically the neural mechanisms of directional hearing. The VAS-earphone procedure requires accurate FETF estimation from free-field measurements and appropriate compensation for the undesirable spectral characteristics of the closed-field earphone sound delivery and measurement systems. Here we describe how specially designed finite-impulse-response (FIR) filters improve these two operations. The coefficients of an FIR filter arc determined using a least-squares error criterion. The least-squares FIR filter is implemented entirely in the time domain and avoids the usual problems with division inherent in a frequency domain approach. The estimation of an FETF by a least-squares FIR filter is veracious since its impulse response can recover signals that were recorded near the eardrum in the free field with a very high fidelity. The correlation coefficient between recorded and recovered time waveforms typically exceeds 0.999. Similarly, least-squares FIR filters prove excellent in compensating closed-field sound systems since comparisons of waveforms delivered by a compensated earphone to their corresponding predistorted signals yield correlation coefficients that exceed 0.99 on average.