Showing papers in "WSEAS Transactions on Signal Processing archive in 2008"

Image compression using neural networks and haar wavelet

Abstract: Wavelet-based image compression provides substantial improvements in picture quality at higher compression ratios. Haar wavelet transform based compression is one of the methods that can be applied for compressing images. An ideal image compression system must yield good quality compressed images with good compression ratio, while maintaining minimal time cost. With Wavelet transform based compression, the quality of compressed images is usually high, and the choice of an ideal compression ratio is difficult to make as it varies depending on the content of the image. Therefore, it is of great advantage to have a system that can determine an optimum compression ratio upon presenting it with an image. We propose that neural networks can be trained to establish the non-linear relationship between the image intensity and its compression ratios in search for an optimum ratio. This paper suggests that a neural network could be trained to recognize an optimum ratio for Haar wavelet compression of an image upon presenting the image to the network. Two neural networks receiving different input image sizes are developed in this work and a comparison between their performances in finding optimum Haar-based compression is presented.

Word and triphone based approaches in continuous speech recognition for Tamil language

Abstract: Building a continuous speech recognizer for the Indian language like Tamil is a challenging task due to the unique inherent features of the language like long and short vowels, lack of aspirated stops, aspirated consonants and many instances of allophones. Stress and accent vary in spoken Tamil language from region to region. But in formal read Tamil speech, stress and accents are ignored. There are three approaches to continuous speech recognition (CSR) based on the sub-word unit viz. word, phoneme and syllable. Like other Indian languages, Tamil is also syllabic in nature. Pronunciation of words and sentences is strictly governed by set of linguistic rules. Many attempts have been made to build continuous speech recognizers for Tamil for small and restricted tasks. However medium and large vocabulary CSR for Tamil is relatively new and not explored. In this paper, the authors have attempted to build a Hidden Markov Model (HMM) based word and triphone acoustic models. The objective of this research is to build a small vocabulary word based and a medium vocabulary triphone based continuous speech recognizers for Tamil language. In this experimentation, a word based Context Independent (CI) acoustic model for 371 unique words and a triphone based Context Dependent (CD) acoustic model for 1700 unique words have been built. In addition to the acoustic models a pronunciation dictionary with 44 base phones and trigram based statistical language model have also been built as integral components of the linguist. These recognizers give very good word accuracy for trained and test sentences read by trained and new speakers.

Design and implementation of digital FIR equiripple notch filter on ECG signal for removal of power line interference

Abstract: Filtering of power line interference is very meaningful in the measurement of biomedical events recording, particularly in the case of recording signals as weak as the ECG. The available filters for power line interference either need a reference channel or regard the frequency as fixed 50/60Hz. Methods of noise reduction have decisive influence on performance of all electro-cardio-graphic (ECG) signal processing systems. This work deals with problems of power line interference reduction. In the literature of the last twenty years several solutions of removal of power line interference on electrocardiogram (ECG) signals can be found. Some analogue and digital approaches to this problem are presented and its properties, advantages and disadvantages are shown. Present paper deals with design and development of digital FIR equiripple filter. The basic ECG has the frequency range from. 5Hz to 100Hz. Artifacts plays the vital role in the processing of the ECG signal. It becomes difficult for the Specialist to diagnose the diseases if the artifacts are present in the ECG signal. In the present work notch filter is designed and applied to the ECG signal containing power line noise. Complete design is performed with FDA tool in the Matlab. The equiripple notch filter designed is having higher order due to which increase in the computational complexity observed. For accessing real time ECG the related instrumentation has been developed in the laboratory. The result shows the ECG signal before filtering and after filtering with their frequency spectrums which clearly indicates the reduction of the power line interference in the ECG signal.

Interference reduction in ECG using digital FIR filters based on rectangular window

Abstract: Coronary heart disease (CHD) is the leading cause of death for both men and women in the all over the world and India too. CHD is caused by a narrowing of the coronary arteries that supply blood to the heart, and often results in a heart attack. Each year, about millions man kind suffers from heart attack. About maximum of those heart attacks are fatal. About half of those deaths occur within 1 hour of the start of symptoms and before the person reaches the hospital. A heart attack is a medical emergency. Hospitalization is required and possibly intensive care. ECG signal is very important signal in the cardiology. Different artifacts are the reason behind the corruption of the signal care should be taken to avoid the interferences in the ECG. The work is in that direction. Present paper deals with the design of the FIR filter using rectangular window. Basically three filters are designed namely low pass filter high pass filter and notch filter. All the filters are cascaded also. These filters are applied on the ECG signal in the real time manner. For the real time application the 711B add-on card has been used. Results clearly indicate that there is noise reduction in the ECG signal. A Comparative Results are Provided in the paper.

Data fusion and topology control in wireless sensor networks

Abstract: The design of large-scale sensor networks interconnecting various sensor nodes has spurred a great deal of interest due to its wide variety of applications. Data fusion is a critical step in designing a wireless sensor network as it handles data acquired by sensory devices. Wireless sensor networks allow distributed sensing and signal processing while collaborating during energy efficient operations. Wireless sensor networks are battery powered; therefore prolonging the network lifetime through an energy aware node organization is highly desirable. The main goal of a topology control scheme in wireless sensor networks is to reduce power consumption in order to extend network lifetime. Our aim is to provide a better understanding of the current research issues in this field. The paper provides a more detailed look at some existing data fusion and topology management algorithms. The most important design issues of data fusion and topology control are also highlighted.

Asymmetric ratio and FCM based salient channel selection for human emotion detection using EEG

Abstract: Electroencephalogram (EEG) is one of the most reliable physiological signals used for detecting the emotional states of human brain. We propose Asymmetric Ratio (AR) based channel selection for human emotion recognition using EEG. Selection of channels reduces the feature size, computational load requirements and robustness of emotions classification. We address this crisis using Asymmetric Variance Ratio (AVR) and Amplitude Asymmetric Ratio (AAR) as new channel selection methods. Using these methods the 28 homogeneous pairs of EEG channels is reduced to 4 and 2 channel pairs respectively. These methods significantly reduce the number of homogeneous pair of channels to be used for emotion detection. This approach is illustrated with 5 distinct emotions (disgust, happy, surprise, sad, and fear) on 63 channels EEG data recorded from 5 healthy subjects. In this study, we used Multi-Resolution Analysis (MRA) based feature extraction the original and reduced set of channels for emotion classification. These approaches were empirically evaluated by using a simple unsupervised classifier, Fuzzy C-Means clustering with variable clusters. The paper concludes by discussing the impact of reduced channels on emotion recognition with larger number of channels and outlining the potential of the new channel selection method.

A new fast forecasting technique using high speed neural networks

Abstract: Forecasting is an important issue for many different applications. In this paper, a new efficient forecasting technique is presented. Such technique is designed by using fast neural networks (FNNs). The new idea relies on performing cross correlation in the frequency domain between the input data and the input weights of neural networks. It is proved mathematically and practically that the number of computation steps required for the proposed fast forecasting technique is less than that needed by conventional neural-based forecasting. Simulation results using MATLAB confirm the theoretical computations. The proposed fast forecasting technique increases the prediction speed and at the same time does not affect the predication accuracy. It is applied professionally for erythemal ultraviolet irradiance prediction.

A locally tuned nonlinear technique for color image enhancement

Abstract: An innovative technique for the enhancement of digital color images captured under extremely non-uniform lighting conditions is proposed in this paper. The key contributions of this technique are adaptive intensity enhancement, contrast enhancement and color restoration. Simultaneous enhancement of extreme dark and bright intensity regions in an image is performed by a specifically designed Locally Tuned Sine Non-Linear (LTSN) function. The intensity of each pixel's magnitude is tuned based on its surrounding pixels to accomplish contrast enhancement. Retrieval of the color information from the enhanced intensity image is achieved by a linear color restoration process which is based on the chromatic information of the input image. Experimental evaluations show that the proposed algorithm can be effectively used for improving the visibility of night time surveillance video sequences with frames having extreme bright and dark regions.

Vision-based distance and area measurement system

Abstract: The objective of this paper is to enable CCD camera for area measuring while recording images simultaneously. Based on an established relationship between pixel number and distance in this paper, we can derive the horizontal and vertical length of a targeted object, and subsequently calculate the area covered by the object. Because of the advantages demonstrated, the proposed system can be used for large-area measurements. For example, we can use this system to measure the size of the gap in the embankments during flooding, or the actual area affected by the landslides. Other applications include the surveying of ecosystems by inspecting how widely spread is a certain type of life form. For places which are difficult or impossible to reach, this system can be particularly useful in performing area measurements. Experiments conducted in this paper have indicated that different shooting distances and angles do not affect the measuring results.

Comparative study of several Fir median hybrid filters for blink noise removal in Electrooculograms

Abstract: The presence of a kind of impulsive noise due to eye blinks is typical during the acquisition of electrooculograms. This paper describes a comparative study of several algorithms used to remove the blink noise in the electroculogram preserving the sharp edges in the signal produced by the so-called saccadic eye movements. Median filters (MF) and several types of Fir Median Hybrid Filters (FMH) have been analyzed. Two types of real electrooculogram register with saccadic movements in controlled position were used to test the performance of the pre-processing filters (sampling rate 20Hz). The filtered signals were later processed with a saccadic eye movement detector algorithm in order to detect changes in the sensitivity and positive predictive value. The results show that neither FMH filters nor WFMH filters produce better results than median filters, in this particular study. The highest averaged values of sensitivity and positive predictive value are obtained by using a median filter of length I=6 samples (S=96.22%, V++=95.42%) and the variant SWFMH of the same length (S=96.27%, V++=91.91%). Although the differences in detection rates are not meaningful between these filters, median filters obtain slightly higher rates of saccades detection than SWFMH, but a reduction in computational burden is obtained by using FHM variants.

Satellite sub-pixel rainfall variability

Abstract: Rain gauge networks are used to calibrate and validate quantitative precipitation estimation (QPE) methods based on remote sensing, which may be used as data sources for hydrologic models. The typical approach is to adjust (calibrate) or compare (validate) the rainfall in the QPE pixel with the rain gauge located within the pixel. The QPE result represents a mean rainfall over the pixel area, whereas the rainfall from the gauge represents a point, although it is normally assumed to represent some area. In most cases the QPE pixel area is millions of square meter in size. We hypothesize that some rain gauge networks in environments similar to this study (i.e., tropical coastal), which provide only one rain gauge per remote sensing pixel, may lead to error when used to calibrate/validate QPE methods, and that consequently these errors may be propagated throughout hydrologic models. The objective of this paper is to describe a ground-truth rain gauge network located in western Puerto Rico which will be available to test our hypothesis. In this paper we discuss observations from the rain gauge network, but do not present any QPE validation results. In addition to being valuable for validating satellite and radar QPE data, the rain gauge network is being used to test and calibrate atmospheric simulation models and to gain a better understanding of the sea breeze effect and its influence on rainfall. In this study, a large number of storms (> 60) were evaluated between August 2006 and August 2008. The area covered by the rain gauge network was limited to a single GOES-12 pixel (4 km × 4 km). Five-minute and total storm rainfall amounts were spatially variable at the sub-pixel scale. The average storm rainfall from 20% of the 120 possible rain gauge-pairs was found to be significantly different at the 5% of significance level, indicating significant rainfall variation at the sub-pixel scale. The average coefficient of determination (r2), describing the goodness of fit of a linear model relating rain gauge pairs, was 0.365, further suggesting a significant degree of variability at the satellite sub-pixel scale. Although there were several different storm types identified (localized, upper westerly trough, tropical easterly wave, tropical westerly trough, cold front and localized with cold front), there did not appear to be any relationship between storm type and the correlation patterns among the gauges.

A hybrid noise cancelling algorithm with secondary path estimation

Abstract: This paper presents a hybrid active noise canceling (HANC) algorithm to overcome the acoustic feedback present in most ANC system, together with an efficient secondary path estimation scheme The HANC system provides a solution of two fundamental problems present in these kind of ANC systems: The first consists in a reduction of the acoustic feedback from the cancellation loudspeaker to the input microphone, using two FIR adaptive filters, one with a feedforward configuration an the other with a feedback adaptive filter configuration To overcome the secondary path modeling problem, a modification of the method proposed by Akhtar is used Computer simulation results are provide to show the noise cancellation and secondary path estimation performance of presented scheme

The detection of gear noise computed by integrating the Fourier and wavelet methods

Abstract: This paper presents a new gearbox noise detection algorithm based on analyzing specific points of vibration signals using the Wavelet Transform. The proposed algorithm is compared with a previously-developed algorithm associated with the Fourier decomposition using Hanning windowing. Simulation carried on real data demonstrate that the WT algorithm achieves a comparable accuracy while having a lower computational cost. This makes the WT algorithm an appropriate candidate for fast processing of noise gear box.

Syllable-based automatic arabic speech recognition in noisy-telephone channel

Abstract: The performance of well-trained speech recognizers using high quality full bandwidth speech data is usually degraded when used in real world environments. In particular, telephone speech recognition is extremely difficult due to the limited bandwidth of transmission channels. In this paper, we concentrate on the telephone recognition of Egyptian Arabic speech using syllables. Arabic spoken digits were described by showing their constructing phonemes, triphones, syllables and words. Speaker-independent hidden markov models (HMMs)-based speech recognition system was designed using Hidden markov model toolkit (HTK). The database used for both training and testing consists from forty-four Egyptian speakers. In clean environment, experiments show that the recognition rate using syllables outperformed the rate obtained using monophones, triphones and words by 2.68%, 1.19% and 1.79% respectively. Also in noisy telephone channel, syllables outperformed the rate obtained using monophones, triphones and words by 2.09%, 1.5% and 0.9% respectively. Comparative experiments have indicated that the use of syllables as acoustic units leads to an improvement in the recognition performance of HMM-based ASR systems in noisy environments. A syllable unit spans a longer time frame, typically three phones, thereby offering a more parsimonious framework for modeling pronunciation variation in spontaneous speech. Moreover, syllable-based recognition has relatively smaller number of used units and runs faster than word-based recognition.

Blind tamper detection in audio using chirp based robust watermarking

Abstract: In this paper, we propose the use of 'chirp coding' for embedding a watermark in audio data without generating any perceptual degradation of audio quality. A binary sequence (the watermark) is derived using energy based features from the audio signal and chirp coding used to embed the watermark in audio data. The chirp coding technique is such that the same watermark can be derived from the original audio signal as well as recovered from the watermarked signal. This not only enables the 'blind' recovery of the watermark, but also provides a solution for deriving two independent extraction processes for the watermark from which it is possible to ensure the authenticity of audio data and any mismatch indicating that the data may have been tampered with. To evaluate the robustness of the proposed scheme, different attacks such as compression, filtering, sampling rate alteration, for example, have been simulated. The results obtained reflect the high robustness of the watermark method used and is effectiveness in detecting any data tampering that may have occurred. For perceptual transparency of the watermark, Perceptual Assessment of Audio Quality (PEAQ ITU-R BS.1387) on Speech Quality Assessment Material (SQAM) has been undertaken and an average of -0.5085 Objective Difference Grade achieved.

The fractal dimension correlated to the bone mineral density

Abstract: Osteoporosis is a condition of decreased bone mass. This leads to fragile bones which are at an increased risk for fractures, more often, it affects postmenopausal women. In this paper we propose a study of osteoporosis with the fractal dimension. After an introduction to the theory and fractal dimension, we use the box counting method for the segmentation of radiographic images, the study of the influence of range size boxes on the fractal dimension will be investigated, and the correlation between a reference dimension and bone mineral density. Other imaging techniques will be given in order to see the results of the application of the method on these types of images.

Ships classification basing on acoustic signatures

Abstract: The paper presents the technique of artificial neural networks used as classifier of hydroacoustic signatures generated by moving ship. The main task of proposed solution is to classify the objects which made the underwater noises. Firstly, the measurements were carried out dynamically by running ship past stationary hydrophones, mounted on tripods 1 m above the sea bottom. Secondly to identify the source of noise the level of vibration were measured on board by accelerometers, which were installed on important components of machinery. On the base of this measurement there was determined the sound pressure level, noise spectra and spectograms, transmission of acoustic energy via the hull into water. More over it was checked by using coherence function that components of underwater noise has its origin in vibrations of ship's mechanisms. Basing on this research it was possible to create the hydroacoustic signature or so called "acoustic portrait" of moving ship. Next during the complex ships' measurements on Polish Navy Test and Evaluation Acoustic Range hydroacoustic noises generated by moving ship were acquired. Basing on these results the classifier of acoustic signatures using artificial neural network was worked out. From the technique of artificial neural networks the Kohonen networks which belongs to group of self organizing networks where chosen to solve the research problem of classification. The choice was caused by some advantages of mentioned kind of neural networks like: they are ideal for finding relationships amongst complex sets of data, they have possibility to self expand the set of answers for new input vectors. To check the correctness of classifier work the research in which the number of right classification for presented and not presented before hydroacoustic signatures were made. Some results of research were presented on this paper. Described method actually is extended and its application is provided as assistant subsystem for hydrolocations systems of Polish Naval ships.

Off-line cursive handwritten Tamil character recognition

Abstract: In spite of several advancements in technologies pertaining to Optical character recognition, handwriting continues to persist as means of documenting information for day-to-day life. The process of segmentation and recognition pose quiets a lot of challenges especially in recognizing cursive handwritten scripts of different languages. The concept proposed is a solution crafted to perform character recognition of hand-written scripts in Tamil, a language having official status in India, Sri Lanka, and Singapore. The approach utilizes discrete Hidden Markov Models (HMMs) for recognizing off-line cursive handwritten Tamil characters. The tolerance of the system is evident as it can overwhelm the complexities arise out of font variations and proves to be flexible and robust. Higher degree of accuracy in results has been obtained with the implementation of this approach on a comprehensive database and the precision of the results demonstrates its application on commercial usage. The methodology promises to present a simple and fast scaffold to construct a full OCR system extended with suitable pre-processing.

Estimating parameters of sinusoids from noisy data using Bayesian inference with simulated annealing

Abstract: In this paper, we consider Bayesian analysis proposed by Bretthorst for estimating parameters of the corrupted signals and incorporate it with a simulated annealing algorithm to obtain a global maximum of the posterior probability density of the parameters. Thus, this analysis offers different approach to find parameter values through a directed, but random, search of the parameter space. For this purpose, we developed a Mathematica code of this Bayesian approach and used it for recovering sinusoids corrupted by random noise. The simulation results support the effectiveness of the method.

FastICA algorithm for the separation of mixed images

Abstract: Independent component analysis is a generative model for observed multivariate data, which are assumed to be mixtures of some unknown latent variables. It is a statistical and computational technique for revealing hidden factors that underlies set of random variable measurements of signals. A common problem faced in the disciplines such as statistics, data analysis, signal processing and neural network is finding a suitable representation of multivariate data. The objective of ICA is to represent a set of multidimensional measurement vectors in a basis where the components are statistically independent. In the present paper we deal with a set of images that are mixed randomly. We apply the principle of uncorrelatedness and minimum entropy to find ICA. The original images are then retrieved using fixed point algorithm known as FastICA algorithm and compared with the original images with the help of estimated error. The outputs from the intermediate steps of algorithm such as PCA, Whitening matrix, Convergence of algorithm and dewhitening matrix are also discussed.

Association-based image retrieval

Abstract: With advances in the computer technology and the World Wide Web there has been an explosion in the amount and complexity of multimedia data that are generated, stored, transmitted, analyzed, and accessed. In order to extract useful information from this huge amount of data, many content-based image retrieval (CBIR) systems have been developed in the last decade. A typical CBIR system captures image features that represent image properties such as color, texture, or shape of objects in the query image and try to retrieve images from the database with similar features. Recent advances in CBIR systems include relevance feedback based interactive systems. The main advantage of CBIR systems with relevance feedback is that these systems take into account the gap between the high-level concepts and low-level features and subjectivity of human perception of visual content. In this paper, we propose a new approach for image storage and retrieval called association-based image retrieval (ABIR). We try to mimic human memory. The human brain stores and retrieves images by association. We use a generalized bi-directional associative memory (GBAM) to store associations between feature vectors. The results of our simulation are presented in the paper.

Slovenian spontaneous speech recognition and acoustic modeling of filled pauses and onomatopoeas

Abstract: This paper is focused on acoustic modeling for spontaneous speech recognition. This topic is still a very challenging task for speech technology research community. The attributes of spontaneous speech can heavily degrade speech recognizer's accuracy and performance. Filled pauses and onomatopoeias present one of such important attributes of spontaneous speech, which can give considerably worse accuracy. Although filled pauses don't carry any semantic information, they are still very important from the modeling perspective. A novel acoustic modeling approach is proposed in this paper, where the filled pauses are modeled using the phonetic broad classes, which corresponds with their acoustic-phonetic properties. The phonetic broad classes are language dependent, and can be defined by an expert or in a data-driven way. The new filled pauses modeling approach is compared with three other implicit filled pauses modeling methods. All experiments were carried out using a context-dependent Hidden Markov Models based speech recognition system. For training and evaluation, the Slovenian BNSI Broadcast News speech and text database was applied. The database contains manually transcribed recordings of TV news shows. The evaluation of the proposed acoustic modeling approach was done on a set of spontaneous speech. The overall best filled pauses acoustic modeling approach improved the speech recognizer's word accuracy for 5.70% relatively in comparison to the baseline system, without influencing the recognition time.

Practical approaches on enhancement and segmentation of trimulus color image with information theory based quantitative measuring

Abstract: Image enhancement and image clustering are two practical implementation approaches for pattern recognition with a variety of engineering applications. In most cases, the actual outcomes of some advanced image processing approaches will directly affect the decision making, such as in target detection and medical diagnosis. Among these approaches, image adaptive contrast stretching is a typical enhancement approach under conditions of improper illumination and unpleasant disturbances, which adapts to the intensity distribution of an image. K-means clustering is a typical segmentation approach to minimize the medium dispersing impact, which produces the distinctive clusters or layers for representing different components of the information being detected. In trimulus color systems, each of three color components takes an independent role along with image processing procedures. To evaluate actual effects of image enhancement and image segmentation, quantitative measures should be taken into account rather than qualitative evaluations exclusively. In this article, quantitative measures for trimulus color systems are proposed instead of the existing gray level ones. Considering the gray level image measures, the corresponding true color RGB component energy, discrete entropy, relative entropy and mutual information are proposed to measure the effectiveness of color image enhancement and segmentation techniques.

A study on ultrasonic signals processing generated from automobile engine block using statistical analysis

Abstract: The development of statistical analysis has played an important part in studying large data that captured from engine block as apart of engine monitoring and diagnose. Within this paper the application of statistical analysis was introduced by utilizing Kurtosis, I-kaz coefficient, and Crest Factor and Skewness parameter. There is potential that these statistical parameters could serve as pattern recognition to identify engine type and characteristic. The study was performed in two stages. The first stage is an experimental process that uses two three-cylinder automobile 845 cc and 850 cc engines and two four-cylinder automobile 1468 cc and 1784 cc engines which run under idle condition. In the second stage, the plots of signal's statistical parameter based on the engine type were done accordingly. As a result, the plot of the statistical parameter against I-kaz coefficient shows an excellent classification pattern. The pattern was useful in determining engine type for signal confirmation and engine fault detection.

Video target tracking by using competitive neural networks

Abstract: A target tracking algorithm able to identify the position and to pursuit moving targets in video digital sequences is proposed in this paper. The proposed approach aims to track moving targets inside the vision field of a digital camera. The position and trajectory of the target are identified by using a neural network presenting competitive learning technique. The winning neuron is trained to approximate to the target and, then, pursuit it. A digital camera provides a sequence of images and the algorithm process those frames in real time tracking the moving target. The algorithm is performed both with black and white and multi-colored images to simulate real world situations. Results show the effectiveness of the proposed algorithm, since the neurons tracked the moving targets even if there is no pre-processing image analysis. Single and multiple moving targets are followed in real time.

Novel statistical approach to blind recovery of earth signal and source wavelet using independent component analysis

Abstract: This paper provides a new statistical approach to blind recovery of both earth signal and source wavelet given only the seismic traces using independent component analysis (ICA) by explicitly exploiting the sparsity of both the reflectivity sequence and the mixing matrix. Our proposed blind seismic deconvolution algorithm consists of three steps. Firstly, a transformation method that maps the seismic trace convolution model into multiple inputs multiple output (MIMO) instantaneous ICA model using zero padding matrices has been proposed. As a result the nonzero elements of the sparse mixing matrix contain the source wavelet. Secondly, whitening the observed seismic trace by incorporating the zero padding matrixes is conducted as a pre-processing step to exploit the sparsity of the mixing matrix. Finally, a novel logistic function that matches the sparsity of reflectivity sequence distribution has been proposed and fitted into the information maximization algorithm to obtain the demixing matrix. Experimental simulations have been accomplished to verify the proposed algorithm performance over conventional ICA algorithms such as Fast ICA and JADE algorithm. The mean square error (MSE) of estimated wavelet and estimated reflectivity sequence shows the improvement of proposed algorithm.

Testing of image segmentation methods

Abstract: Digital image segmentation is broadly used in various image processing tasks. A large amount of image segmentation methods gives rise to the problem of of method's choice, most adequate for practical purposes. In this paper, we develop an approach which allows quantitative and qualitative estimation of segmentation programs. It consists in modeling both difficult and typical situations in image segmentation tasks using special sets of artificial test images. The description of test images and testing procedures are given. Our approach clears up specific features and applicability limits of four segmentation methods under examination.

Fault characterisation and classification using wavelet and fast Fourier transforms

Abstract: In order to improve the power quality maintenance and reliability of power supply, different types of faults on the transmission line namely: open-circuit (OC), short-circuit (SC), high impedance faults (HIF) and the fault caused by direct lightning strike (LS) have been investigated in this paper. The disturbances have been modelled and simulated using a well-known transient simulation tool - Alternative Transient Program/ Electromagnetic Transient Program (ATP/EMTP) and the resulting data are then imported into MATLAB for the investigation on the traveling wave (TW) reflection pattern and harmonic behaviour. Study on the characteristics of the faults in terms of their corresponding frequency spectrum, the polarities of the incident-wave and reflected-wave has been performed and the possibility to differentiate the type of fault is explored. For this purpose, the fault on the wave has been created at the moment when the voltage signal reaches its peak and also when it is close to zero. Both, Wavelet Transform (WT) and Fast Fourier Transform (FFT) methods have been used to analyze the transient signals generated by the fault. Model of the network used in this study is taken from [1]-[2].

Object-oriented analysis applied to high resolution satellite data

Abstract: The aim of this contribute is to examine an application of Object Oriented Image Analysis on very high resolution data, on Ikonos images - multispectral and panchromatic - of Bagnara Calabra, in the province of Reggio Calabria Our objectives are to show as an automatic analysis as we implemented in a unitary package for segmentation and classification Neuro Fuzzy - with a minimal manual participation - can get a good classification also in presence of high and very high resolution data of small cities, where higher is an error possibility

Computationally efficient algorithm for fuzzy rule-based enhancement on JPEG compressed color images

Abstract: In the past few years the resolution of images increased and the requirement for large storage space and fast process, directly in the compressed domain, becomes essential. Fuzzy rule-based contrast enhancement, is a well-known rather simple approach with good visual results. As any fuzzy algorithm, it is by default nonlinear, thus not straightforward applicable on the JPEG bitstream data - zig-zag ordered quantized DCT (Discrete Cosine Transform) coefficients. Because of their nonlinear nature the fuzzy techniques don't have yet a well-defined strategy for their implementation in the compressed domain. In this paper, we propose an implementation strategy suitable for single input - single output Takagi-Sugeno fuzzy systems with trapezoidal shaped input membership function, directly in the JPEG compressed domain. The fuzzy sets parameters are adaptively chosen by analyzing the histogram of the image data in the compressed domain, in order to optimally enhance the image contrast. The fuzzy rule-based algorithm requires some threshold comparisons, for which an adaptive implementation, taking into account the frequency content of each block in the compress domain JPEG image is proposed. This guarantees the minimal error implementation at minimum computational cost.