By Topic

Signal Processing and Information Technology (ISSPIT), 2009 IEEE International Symposium on

Date 14-17 Dec. 2009

Filter Results

Displaying Results 1 - 25 of 118
  • [Front cover]

    Page(s): c1
    Save to Project icon | Request Permissions | PDF file iconPDF (3662 KB)  
    Freely Available from IEEE
  • Welcome message from the technical program chair

    Page(s): 1
    Save to Project icon | Request Permissions | PDF file iconPDF (2316 KB)  
    Freely Available from IEEE
  • Organizers

    Save to Project icon | Request Permissions | PDF file iconPDF (2122 KB)  
    Freely Available from IEEE
  • Program committee members

    Page(s): 1 - 2
    Save to Project icon | Request Permissions | PDF file iconPDF (2224 KB)  
    Freely Available from IEEE
  • Table of content

    Page(s): 1 - 5
    Save to Project icon | Request Permissions | PDF file iconPDF (241 KB)  
    Freely Available from IEEE
  • Quick invariant signature extraction from binary images

    Page(s): 172 - 177
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (2319 KB) |  | HTML iconHTML  

    It is shown how basic geometric notions can be used to extract an image signature independently of position, orientation and size. Simple primitives as lengths and slopes remain invariant by affine 2D similarity transformations. They can easily be used to define the invariant signature of an image. Contrary to the previous work in this area, images can be directly analyzed. This means that the extraction of interest points of the image is avoided. The method remains formal and no estimation or compression is needed. It is formally demonstrated that 100% of transformations are taken into consideration and that the signature of the image is totally invariant. The Quick Invariant Signature (QIS) extraction is a formal and fast method. It can be used either, only for signature extraction, or be integrated into a neural architecture for both extraction and classification. Unusual invariances such as cylindrical translation or toric translation are also defined by QIS. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Assessment of CELP codecs quality in multi-lingual environment

    Page(s): 55 - 60
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (312 KB) |  | HTML iconHTML  

    This paper investigates the performance of CELP speech codecs over different languages. The English language has had a dominating influence in the advance of telecommunications. With many of the major developments coming from primarily English speaking areas there is the risk that these advances may not be linguistically robust. It is noted that quality of a speech produced by voice codecs mainly is assessed using samples of English language. Some investigation show that language influence to codecs performance could be noticed. In order to judge the performance of the most popular CELP voice codecs (Speex and AMR), we encoded and decoded the speech samples from three different languages: English, Arabic and Lithuanian. The quality of transformed speech signals was estimated using two quality estimation algorithms 3SQM (ITU recommendations P.563) and PESQ (ITU recommendations P.862). The experiments results showed quality bias toward the English language-the scores were higher and the performance was more stable. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • A dynamic approach to reliable mobile agents systems using group communication services.

    Page(s): 71 - 76
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (2292 KB) |  | HTML iconHTML  

    Mobile agent is a process that can transport its state from one environment to another, with its data intact, and be capable of performing appropriately in the new environment. Fault tolerance support of mobile agent execution is essential for achieving a high and reliable performance for the computing process executed by the agent in distributed systems. Most of existing mobile agent systems considers checkpointing or replication as a mechanism in achieving the fault tolerant property. In this paper we present new protocol which employs the benefits gained from combing both mechanisms to achieve reliable mobile agent execution. Our approach uses group communication services to avail different essential issues such as agent's synchronization to facilitate the implementation the protocol. The proposed approach is dynamic in the sense that it allows a flexible membership mechanism to join or leave a mobile agent groups used in achieving the reliable execution. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Robust two-stage lip tracker

    Page(s): 265 - 270
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (619 KB) |  | HTML iconHTML  

    We propose a novel two-stage lip tracking method. A model-based inverse compositional (MIC) feature tracker is first used to track the outer lip, and a fast block matching algorithm is then used to track the inner lip. The proposed two-stage lip tracker outperforms single stage lip trackers such as AAM lip tracker and MIC lip tracker in terms of lower RMS fitting error and faster lip fitting speed. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • An E-learning model based on dynamic qualitative constraint networks

    Page(s): 77 - 82
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (3097 KB) |  | HTML iconHTML  

    This paper presents a new approach to model the core part of an e-learning process. The formal modeling we use relies on Allen's Interval Algebra to build the qualitative constraints networks. With the corresponding intervals, we could represent the conditional sequence of activities the learner is expected to conduct in order to complete a particular course. We propose temporal representation of learning activities and we build the constraints satisfaction problem network according to the activation constraints of learning activities and the activation conditions. The algorithm demonstrates the consistency of the corresponding network and provides the learners with learning scenarios they may take during a particular material learning process. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Generic multi-document summarization using cluster refinement and NMF

    Page(s): 65 - 70
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (2290 KB) |  | HTML iconHTML  

    In this paper, a generic summarization method that uses cluster refinement and NMF is introduced to extract meaningful sentences from documents. The proposed method uses cluster refinement to improve the quality of document clustering since it helps us to remove dissimilarity information easily and avoid biased inherent semantics of documents to be reflected in clusters by NMF. In addition, it uses the weighted semantic variable to select meaningful sentences because the extracted sentences are well covered with the major topics of document. The experimental results demonstrate that the proposed method has better performance than other methods that use the other methods. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Online Social Networks - An interface requirements analysis

    Page(s): 550 - 556
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (3226 KB) |  | HTML iconHTML  

    Online Social Networking (OSN) Sites are virtual places that cater to a specific population in which people of similar interest gather to communicate, share and discuss ideas. Many researchers have studied the effects of these networks and most have inferred that they foster relationship building and communications among those involved. This research is a study aimed at identifying the interface features most desired by users of well known Online Social Network (OSN's). The authors have undertaken an Empirical. Investigation from two demographic subsets of people from United Arab Emirates. The outcome of this study could be of importance to the design of innovative Online Social Networks. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Dynamic fuzzy clustering using Harmony Search with application to image segmentation

    Page(s): 538 - 543
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1361 KB) |  | HTML iconHTML  

    In this paper, a new dynamic clustering approach based on the harmony search algorithm (HS) called DCHS is proposed. In this algorithm, the capability of standard HS is modified to automatically evolve the appropriate number of clusters as well as the locations of cluster centers. By incorporating the concept of variable length in each harmony memory vector, DCHS is able to encode variable numbers of candidate cluster centers at each iteration. The PBMF cluster validity index is used as an objective function to validate the clustering result obtained from each harmony memory vector. The proposed approach has been applied onto well known natural images and experimental results show that DCHS is able to find the appropriate number of clusters and locations of cluster centers. This approach has also been compared with other metaheuristic dynamic clustering techniques and has shown to be very promising. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Time-weighted quantitative testing of image segmentation with a genetic algorithm

    Page(s): 271 - 276
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (1641 KB) |  | HTML iconHTML  

    Automatic parameter selection for image segmentation is accelerated by means of a genetic algorithm (GA). Issues remain in selecting the population size, number of generations, and termination of the search. As evaluation time and subsequent image batch-processing time, when the algorithm is applied in practice, are important considerations, this paper introduces a time factor into the GA cost function. It is found that this procedure while preserving the GA solution also improves interpretation and parameter selection. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Source geography estimation for web pages

    Page(s): 457 - 462
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (368 KB) |  | HTML iconHTML  

    The problem of inferring geographical information associated to web pages and identifying the geographic scope of their content is gaining increasing attention. However, geographic scope is a concept that can be interpreted in many different ways, ranging from the expected target scope of a specific content to the country where the content originated. The latter, in particular, albeit difficult to address, is of great importance for many reasons, such as, for example, market inquiries or anytime estimates on content production in specific countries are needed. Search engines may also be affected by the knowledge of the various kinds of geographic scopes, to better tune their responses to queries, e.g. according to (but not restricted to) the geographic proximity with the user location. However that information is rarely available and must be inferred in the vast majority of the cases. In this paper we propose a technique, grounded into the machine learning theory, to estimate source geography of web pages by means of a classifier learned on a specially constructed training set. The training set, consisting of a number of features extracted from web pages and the corresponding source-geography label (i.e. the country of origin of the web page) is automatically built by exploiting the wide number of pages with contents licensed under a localized Creative Commons (CC) license. The model thus learned is then used to classify unlabeled records and our tests showed a mean accuracy of 81% with a standard deviation of 0:9. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Flooding Zone Control Protocol (FZCP): enhancing the reliability of real-time multimedia delivery in WSNs

    Page(s): 451 - 456
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (448 KB) |  | HTML iconHTML  

    The flooding zone initialization protocol (FZIP) was proposed as a mechanism to provide power efficient flooding for real-time multimedia data over wireless sensor networks (WSNs). FZIP can initialize different FZ sizes with different performance levels (i.e. loss rate, latency, and overhead). Increasing the FZ size increases redundancy which in turn reduces packet loss but consumes extra power overhead. However, how to choose a suitable FZ size poses a tradeoff between packet loss rate and power efficiency under different network sizes, densities, and radio channel conditions. The static FZ size estimation under dynamic WSNs' conditions leads to unnecessary power overhead or fail to deliver good quality resulting in high packet loss rate. In this paper, we propose the flooding zone control protocol (FZCP) to overcome this problem. FZCP monitors the incoming multimedia packets to detect performance deterioration and change the FZ size (increase/decrease) accordingly. Simulation results show that FZCP enhances the flooding performance and delivers and maintains good quality of real-time multimedia sessions with low energy overhead. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Improving the performance of the instantaneous Blind audio Source Separation algorithms

    Page(s): 519 - 526
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (6104 KB) |  | HTML iconHTML  

    Several algorithms for instantaneous blind source separation (BSS) have been introduced in the past years. The performance of these algorithms needs to be evaluated and assessed to study their merits and choose the best of them for a given application. In this paper, a new adaptive approach is presented to evaluate different blind source separation algorithms. In this new approach, three new evaluation metrics are added. The first metric is the minimum number of samples required for a successful separation process. The second metric is the time needed to complete the separation process. The third metric is the number of sources that the BSS algorithm can separate from their mixtures. The new approach is used to compare three different blind source separation algorithms. These algorithms are: kurtosis, negentropy, and the maximum likelihood. Since the evaluation of a BSS technique is application-dependent, we are using the same application (separation of audio sources) to evaluate each of these BSS algorithms. The comparison, between the three algorithms, shows that the maximum likelihood has the best performance and the kurtosis is the faster. This motivates us to develop a new hybrid approach that combines the two algorithms to gain the benefits from both algorithms. In this new algorithm we start with the maximum likelihood (ML) algorithm to find the separation matrix and then tune this matrix by the kurtosis algorithm. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Reversible blind watermarking for medical images based on wavelet histogram shifting

    Page(s): 31 - 36
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (609 KB) |  | HTML iconHTML  

    This paper proposes a blind reversible watermarking approach for medical images based on histogram shifting in wavelet domain. An integer wavelet transform is applied to map the integer host image components to integer wavelet coefficients. The watermark information is inserted into the high frequency subband regions of the transformed image. According to the capacity required for the watermark, two thresholds, T1 and T2, are selected, one in the begging part and the other in the end part of the histogram of the high frequency subbands of the transformed image. Two zero-points, Z1 and Z2 are also created by properly shifting the beginning and the end parts of the histogram. The part of the histogram located between the two thresholds remains unchanged. The binary watermark data are inserted in the thresholds and zero-points locations. The high PSNR (above 53 dB) obtained for several watermarked medical images, indicates imperceptibility of the approach. Experimental results also show superiority of the proposed approach in compare to some other methods that are based on histogram shifting in spatial as well as integer wavelet domains. Enabling lossless reconstruction of both watermark and host image, beside providing the high quality for the watermarked image, make the proposed approach attractive for medical image watermarking applications. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Intuitive fuzzy c-means algorithm

    Page(s): 83 - 88
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (2328 KB) |  | HTML iconHTML  

    Fuzzy C-means (FCM) is one of the most widely used clustering algorithms and assigns memberships to which are inversely related to the relative distance to the point prototypes that are cluster centers in the FCM model. In order to overcome the problem of outliers in data, several models including possibilistic C-means (PCM) and possibilistic-fuzzy C-means (PFCM) models have been proposed. A new model called intuitive fuzzy C-means (IFCM) model is proposed in this paper. In IFCM, a new measurement called intuition level is introduced so that the intuition level helps to alleviate the effect of noise. Several numerical examples are used for experiments to compare the clustering performance of IFCM with those of FCM, PCM, and PFCM. Results show that IFCM compares favorably to the FCM, PCM, and PFCM models. Since IFCM produces cluster prototypes less sensitive to outliers and to the selection of involved parameters than the other algorithms, IFCM is a good candidate for data clustering problems. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Classification of video data using Centroid Neural Network

    Page(s): 408 - 411
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (2080 KB) |  | HTML iconHTML  

    A classification method of video data using centroid neural network is proposed in this paper. The CNN algorithm is used for clustering the MPEG video data. In comparison with other conventional algorithms, The CNN requires neither a predetermined schedule for learning gain nor the total number of iterations for clustering. It always converges to sub-optimal solutions while conventional algorithms such as SOM may give unstable results depending on the initial learning gains and the total number of iterations. Experiments and results on several MPEG video data sets demonstrate that the classification model employing the CNN can archive improvements in terms of false alarm rate (FAR) over the models using the conventional k-means and SOM algorithms. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Speech recognition system for embedded real-time applications

    Page(s): 118 - 122
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (500 KB) |  | HTML iconHTML  

    In this paper, a hardware/software co-processing speech recognizer for embedded applications is proposed. The system mainly consists of a softcore processor and a hardware accelerator. The accelerator is responsible for GMM emission probability calculation, which is the major computational bottleneck. To alleviate the memory bandwidth issue, the hardware accelerator uses double-buffering, which allows parallel operation of data retrieval and GMM computation. The proposed accelerator is synthesized on an Altera Stratix II FPGA device together with a Nios II softcore processor running at 100 MHz. The proposed system is compared with a pure software-based system using test utterances from the Resource Management (RMI) corpus. For a speech utterance length of 2.49 s, the decoding time reduces from 6.64 s to 2.48 s. The real-time factor improves from 2.67 to 1.00. The word accuracy rate of the proposed system on the RM corpus is 93:42%. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Partitioned Feature-based Classifier model

    Page(s): 412 - 417
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (5397 KB)  

    The Partitioned Feature-based Classifier (PFC) is proposed in this paper. PFC does not use entire feature vectors extracted from the original data at once to classify each datum, but use only groups of features related to each feature vector to classify data separately. In the training stage, the contribution rate calculated from each feature vector group is drawn throughout the accuracy of each feature vector group and then, in the testing stage, the final classification result is obtained by applying weights corresponding to the contribution rate of each feature vector group. The proposed PFC algorithm is applied to two audio data classification problems, a speech/music data classification problem and a music genre classification problem. The results demonstrate that conventional clustering algorithms can improve their classification accuracy when the proposed PFC model is used with them. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • Full text access may be available. Click article title to sign in or learn about subscription options.
  • A new psycho-acoustic model For MPEG1 layer 3 coder using A dynamic Gammachirp wavelet

    Page(s): 123 - 128
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (8255 KB) |  | HTML iconHTML  

    In this paper, we present an approach for the psycho-acoustic analysis of musical signals based on the gamma chirp wavelet packet transform following the model used in the standard MPEG-1 audio layer 3. The proposed method mimics the multiresolution properties of the human ear closer than other techniques and it includes simultaneous and temporal auditory masking. Its important characteristic is to propose an analysis of the frequency bands that come closer to the critical bands of the ear. This study shows the best performance of the gamma chirp coder. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.
  • High throughput multiple-precision GCD on the CUDA architecture

    Page(s): 507 - 512
    Save to Project icon | Request Permissions | Click to expandQuick Abstract | PDF file iconPDF (3967 KB) |  | HTML iconHTML  

    Investigation of the cryptanalytic strength of RSA cryptography requires computing many GCDs of two long integers (e.g., of length 1024 bits). This paper presents a high throughput parallel algorithm to perform many GCD computations concurrently on a GPU based on the CUDA architecture. The experiments with an NVIDIA GeForce GTX285 GPU and a single core of 3.0 GHz Intel Core2 Duo E6850 CPU show that the proposed GPU algorithm runs 11.3 times faster than the corresponding CPU algorithm. View full abstract»

    Full text access may be available. Click article title to sign in or learn about subscription options.