Pub Date : 2006-04-17DOI: 10.1109/SIU.2006.1659766
H. Altınçay, C. Ergun, T. Çiloglu
Following the anticipation that some of the MPEG-7 audio descriptors hold glottal information differing than those MFCCs hold, possible contribution of MPEG-7 descriptors to speaker verification has been investigated. Both feature level and score level fusion of MFCCs and MPEG-7 descriptors have been studied. Results indicate improvements up to 18 % compared to those obtained by using MFCCs alone; a justification of the anticipation and a novel indication to the community
{"title":"Fusion of MFCC and MPEG-7 Attributes for Speaker Verification","authors":"H. Altınçay, C. Ergun, T. Çiloglu","doi":"10.1109/SIU.2006.1659766","DOIUrl":"https://doi.org/10.1109/SIU.2006.1659766","url":null,"abstract":"Following the anticipation that some of the MPEG-7 audio descriptors hold glottal information differing than those MFCCs hold, possible contribution of MPEG-7 descriptors to speaker verification has been investigated. Both feature level and score level fusion of MFCCs and MPEG-7 descriptors have been studied. Results indicate improvements up to 18 % compared to those obtained by using MFCCs alone; a justification of the anticipation and a novel indication to the community","PeriodicalId":415037,"journal":{"name":"2006 IEEE 14th Signal Processing and Communications Applications","volume":"9 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116567374","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2006-04-17DOI: 10.1109/SIU.2006.1659698
S. Altunay, Z. Telatar, O. Eroğul, E. Aydur
Uroflowmetry is a measuring method, which provides numerical and graphical information about patient's lower urinary tract dynamics by measuring and plotting the rate of change of voided urine volume. The main purpose of the study is to evaluate uroflowmetric data using artificial neural networks (ANN) and provide a pre-diagnostic result for urology specialists. The ANN is trained using back-propagation method and the inputs of ANN are the results of a special feature extraction algorithm, which is designed with the suggestions of urology specialists. System's success is monitored with a set of data, which was already diagnosed by specialists. The outputs of ANN are classified into three groups, namely, "healthy", "possible pathologic" and "pathologic"
{"title":"Interpretation of Uroflow Graphs with Artificial Neural Networks","authors":"S. Altunay, Z. Telatar, O. Eroğul, E. Aydur","doi":"10.1109/SIU.2006.1659698","DOIUrl":"https://doi.org/10.1109/SIU.2006.1659698","url":null,"abstract":"Uroflowmetry is a measuring method, which provides numerical and graphical information about patient's lower urinary tract dynamics by measuring and plotting the rate of change of voided urine volume. The main purpose of the study is to evaluate uroflowmetric data using artificial neural networks (ANN) and provide a pre-diagnostic result for urology specialists. The ANN is trained using back-propagation method and the inputs of ANN are the results of a special feature extraction algorithm, which is designed with the suggestions of urology specialists. System's success is monitored with a set of data, which was already diagnosed by specialists. The outputs of ANN are classified into three groups, namely, \"healthy\", \"possible pathologic\" and \"pathologic\"","PeriodicalId":415037,"journal":{"name":"2006 IEEE 14th Signal Processing and Communications Applications","volume":"13 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122191491","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2006-04-17DOI: 10.1109/SIU.2006.1659794
I. Ozbek, M. Demirekler
This study is about tracking of formant frequencies using Kalman filtering. Assuming that the formant frequencies are changing in time slowly, it is possible to model their behaviors as outputs of a dynamic system and then track them using Kalman filtering approach. Tracking system considers also the orders and possible intervals of each formant frequency
{"title":"Tracking of Speech Formant Frequencies","authors":"I. Ozbek, M. Demirekler","doi":"10.1109/SIU.2006.1659794","DOIUrl":"https://doi.org/10.1109/SIU.2006.1659794","url":null,"abstract":"This study is about tracking of formant frequencies using Kalman filtering. Assuming that the formant frequencies are changing in time slowly, it is possible to model their behaviors as outputs of a dynamic system and then track them using Kalman filtering approach. Tracking system considers also the orders and possible intervals of each formant frequency","PeriodicalId":415037,"journal":{"name":"2006 IEEE 14th Signal Processing and Communications Applications","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122315891","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2006-04-17DOI: 10.1109/SIU.2006.1659817
T. E. Tuncer
Interpolation is an important problem of signal processing. Even though there are several methods for interpolation, it is still an open problem. In this paper, we propose a new method for interpolation with certain advantages compared to the previous methods. The proposed method is based on the least squares error optimum design of the interpolating filter. Interpolating filter is chosen as the Kaiser filter since it can be configured in a variety of shapes by the appropriate choice of cut-off and shape parameters. Proposed method can be seen as the generalization of the spline interpolator by employing a Kaiser filter. It is shown that the proposed interpolator performs much better than any of its competitors when the signal is at least approximately bandlimited
{"title":"A New Method For Signal Interpolation","authors":"T. E. Tuncer","doi":"10.1109/SIU.2006.1659817","DOIUrl":"https://doi.org/10.1109/SIU.2006.1659817","url":null,"abstract":"Interpolation is an important problem of signal processing. Even though there are several methods for interpolation, it is still an open problem. In this paper, we propose a new method for interpolation with certain advantages compared to the previous methods. The proposed method is based on the least squares error optimum design of the interpolating filter. Interpolating filter is chosen as the Kaiser filter since it can be configured in a variety of shapes by the appropriate choice of cut-off and shape parameters. Proposed method can be seen as the generalization of the spline interpolator by employing a Kaiser filter. It is shown that the proposed interpolator performs much better than any of its competitors when the signal is at least approximately bandlimited","PeriodicalId":415037,"journal":{"name":"2006 IEEE 14th Signal Processing and Communications Applications","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129880959","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2006-04-17DOI: 10.1109/SIU.2006.1659787
C. Demirkir, B. Sankur
Object detection in still images is one of the common problems which is needed to be solved in a robust and reliable manner. Main focus on this work is the designing of classifiers based on Haar like simple features to obtain a good and efficient detection performance. This problem corresponds to the so called feature selection problem which is common in the pattern classifier systems. Classifiers used to detect objects are based on the simple Haar like features and these features are selected using systematic and general evolutionary based algorithm. The objective is to build a set of classifiers which respond stronger to the features present in object patterns than to non-object patterns, thereby improving the class discrimination between these two classes. This approach combines the classifier design with feature selection by using a genetic algorithm (GA). In the feature selection part of the algorithm a GA algorithm which the Haar features are encoded using their parameters in a single chromosome and optimized using genetic operators. During optimization the features which show similar characteristics in the parameter space are selected using a cluster based partitioning algorithm and thereby redundancy in the features is eliminated and a more compact Haar feature set can be obtained. Performances of the resulting chromosomes are measured using a fitness measure which is based on the separation of the two classes samples over a validation set. The resulting object detection structure is tested for near frontal face images in the cluttered background images
{"title":"Object Detection Using Haar Feature Selection Optimization","authors":"C. Demirkir, B. Sankur","doi":"10.1109/SIU.2006.1659787","DOIUrl":"https://doi.org/10.1109/SIU.2006.1659787","url":null,"abstract":"Object detection in still images is one of the common problems which is needed to be solved in a robust and reliable manner. Main focus on this work is the designing of classifiers based on Haar like simple features to obtain a good and efficient detection performance. This problem corresponds to the so called feature selection problem which is common in the pattern classifier systems. Classifiers used to detect objects are based on the simple Haar like features and these features are selected using systematic and general evolutionary based algorithm. The objective is to build a set of classifiers which respond stronger to the features present in object patterns than to non-object patterns, thereby improving the class discrimination between these two classes. This approach combines the classifier design with feature selection by using a genetic algorithm (GA). In the feature selection part of the algorithm a GA algorithm which the Haar features are encoded using their parameters in a single chromosome and optimized using genetic operators. During optimization the features which show similar characteristics in the parameter space are selected using a cluster based partitioning algorithm and thereby redundancy in the features is eliminated and a more compact Haar feature set can be obtained. Performances of the resulting chromosomes are measured using a fitness measure which is based on the separation of the two classes samples over a validation set. The resulting object detection structure is tested for near frontal face images in the cluttered background images","PeriodicalId":415037,"journal":{"name":"2006 IEEE 14th Signal Processing and Communications Applications","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129978537","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2006-04-17DOI: 10.1109/SIU.2006.1659832
A. Bayram, Erol Yildirim, T. Demiralp, A. Ademoglu
The highest temporal resolution, which is crucial for temporal localization of intracerebral activities, is achieved by ERP, but spatial resolution of scalp topography is low. To overcome the limitation of scalp topography, several current-density estimation techniques were developed whose goal is to find the locations of the three-dimensional (3D) intracerebral activities by solving an inverse problem (such as LORETA). However, scalp topologies constituted by multiple sources which makes the inverse problem complicated. The overall objective of this work is to isolate spatial frequency components of scalp topography by 2-D wavelet transform and to interpret spatial frequency formation via corresponding current-density estimations. Moreover, by achieving less complex scalp maps, obstacle of the inverse problem due to the multiple sources might be lessen. At the first step, main topologies of ERP recordings were investigated by hierarchical clustering algorithm. Secondly, different spatial frequencies of these main topologies were separated by 2-D wavelet transform. Finally, main topological maps and topographic maps of different spatial frequencies derived from them were used to find corresponding cortical activities by LORETA. Assessment of our spatial analysis results was made according to the current density estimation results
{"title":"Spatial Frequency Components of to the Event Related Brain Potentials (ERP)","authors":"A. Bayram, Erol Yildirim, T. Demiralp, A. Ademoglu","doi":"10.1109/SIU.2006.1659832","DOIUrl":"https://doi.org/10.1109/SIU.2006.1659832","url":null,"abstract":"The highest temporal resolution, which is crucial for temporal localization of intracerebral activities, is achieved by ERP, but spatial resolution of scalp topography is low. To overcome the limitation of scalp topography, several current-density estimation techniques were developed whose goal is to find the locations of the three-dimensional (3D) intracerebral activities by solving an inverse problem (such as LORETA). However, scalp topologies constituted by multiple sources which makes the inverse problem complicated. The overall objective of this work is to isolate spatial frequency components of scalp topography by 2-D wavelet transform and to interpret spatial frequency formation via corresponding current-density estimations. Moreover, by achieving less complex scalp maps, obstacle of the inverse problem due to the multiple sources might be lessen. At the first step, main topologies of ERP recordings were investigated by hierarchical clustering algorithm. Secondly, different spatial frequencies of these main topologies were separated by 2-D wavelet transform. Finally, main topological maps and topographic maps of different spatial frequencies derived from them were used to find corresponding cortical activities by LORETA. Assessment of our spatial analysis results was made according to the current density estimation results","PeriodicalId":415037,"journal":{"name":"2006 IEEE 14th Signal Processing and Communications Applications","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131016708","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2006-04-17DOI: 10.1109/SIU.2006.1659750
N. H. Ekmekci, M. Ozer
Voltage-gated ion channels are of great importance in the generation and propagation of electrical signals in the excitable membranes. In this study, we introduce the stochastic version of the Hodgkin-Huxley formalism and investigate the effect of channel noise on neuronal dynamic behaviors based on a model with the Gaussian noise. We show that the channel noise may result in a spiking activity even in the absence of any stimulus for small membrane patches, and that the spontaneous firing dynamics follows more regular patterns when the membrane patch becomes smaller. It is also shown that the stochastic model converges to the deterministic model for very large membrane patches and regularity pattern of fired spikes exhibit resonance behaviour depending on the stimuli strength and the membrane patch area
{"title":"Effect of DC Stimuli on Neuronal Dynamics","authors":"N. H. Ekmekci, M. Ozer","doi":"10.1109/SIU.2006.1659750","DOIUrl":"https://doi.org/10.1109/SIU.2006.1659750","url":null,"abstract":"Voltage-gated ion channels are of great importance in the generation and propagation of electrical signals in the excitable membranes. In this study, we introduce the stochastic version of the Hodgkin-Huxley formalism and investigate the effect of channel noise on neuronal dynamic behaviors based on a model with the Gaussian noise. We show that the channel noise may result in a spiking activity even in the absence of any stimulus for small membrane patches, and that the spontaneous firing dynamics follows more regular patterns when the membrane patch becomes smaller. It is also shown that the stochastic model converges to the deterministic model for very large membrane patches and regularity pattern of fired spikes exhibit resonance behaviour depending on the stimuli strength and the membrane patch area","PeriodicalId":415037,"journal":{"name":"2006 IEEE 14th Signal Processing and Communications Applications","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133665407","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2006-04-17DOI: 10.1109/SIU.2006.1659843
ilkay Ulusoy, C. Bishop
Object recognition from 2D images is a highly interesting problem. The final goal is to have a system which can recognize thousands of different categories like human beings do. However, hand labelling the 2D training images in order to segment the foreground (object) from the background is a very tedious job. Because of this reason, in recent years, intelligent systems which can learn object categories from unlabelled image sets have been introduced. In this case, an image is labelled by the objects which are present in the image but the objects are not segmented in the image. The main problem in this case is that the object and the background are used altogether in such unsupervised systems and segmentation must be performed by the system itself. Automatic Relevance Determination (ARD ) is a method which will be investigated in this study in order to segment foreground and background in an unsupervised object category learning system.
{"title":"Automatic Relevance Determination for the Estimation of Relevant Features for Object Recognition","authors":"ilkay Ulusoy, C. Bishop","doi":"10.1109/SIU.2006.1659843","DOIUrl":"https://doi.org/10.1109/SIU.2006.1659843","url":null,"abstract":"Object recognition from 2D images is a highly interesting problem. The final goal is to have a system which can recognize thousands of different categories like human beings do. However, hand labelling the 2D training images in order to segment the foreground (object) from the background is a very tedious job. Because of this reason, in recent years, intelligent systems which can learn object categories from unlabelled image sets have been introduced. In this case, an image is labelled by the objects which are present in the image but the objects are not segmented in the image. The main problem in this case is that the object and the background are used altogether in such unsupervised systems and segmentation must be performed by the system itself. Automatic Relevance Determination (ARD ) is a method which will be investigated in this study in order to segment foreground and background in an unsupervised object category learning system.","PeriodicalId":415037,"journal":{"name":"2006 IEEE 14th Signal Processing and Communications Applications","volume":"23 5","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132152922","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2006-04-17DOI: 10.1109/SIU.2006.1659822
Asanterabi Malima, E. Ozgur, M. Çetin
We propose a fast algorithm for automatically recognizing a limited set of gestures from hand images for a robot control application. Hand gesture recognition is a challenging problem in its general form. We consider a fixed set of manual commands and a reasonably structured environment, and develop a simple, yet effective, procedure for gesture recognition. Our approach contains steps for segmenting the hand region, locating the fingers, and finally classifying the gesture. The algorithm is invariant to translation, rotation, and scale of the hand. We demonstrate the effectiveness of the technique on real imagery
{"title":"A Fast Algorithm for Vision-Based Hand Gesture Recognition for Robot Control","authors":"Asanterabi Malima, E. Ozgur, M. Çetin","doi":"10.1109/SIU.2006.1659822","DOIUrl":"https://doi.org/10.1109/SIU.2006.1659822","url":null,"abstract":"We propose a fast algorithm for automatically recognizing a limited set of gestures from hand images for a robot control application. Hand gesture recognition is a challenging problem in its general form. We consider a fixed set of manual commands and a reasonably structured environment, and develop a simple, yet effective, procedure for gesture recognition. Our approach contains steps for segmenting the hand region, locating the fingers, and finally classifying the gesture. The algorithm is invariant to translation, rotation, and scale of the hand. We demonstrate the effectiveness of the technique on real imagery","PeriodicalId":415037,"journal":{"name":"2006 IEEE 14th Signal Processing and Communications Applications","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132571068","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2006-04-17DOI: 10.1109/SIU.2006.1659793
O. Alay, G. Akar
In H.264 video coding standard, in order to increase the coding efficiency, several new techniques such as spatial intra prediction, integer transformation have been introduced. However these also increase the computational complexity drastically. In this paper, we propose a new fast intra prediction algorithm to be used instead of the full search algorithm to choose the best mode for spatial intra prediction. Experimental results show that, our algorithm achieves 52% computation reduction on the average while maintaining similar PSNR and an average bitrate increase of 1.5%
{"title":"Fast Mode Decision for Intra Prediction in H.264","authors":"O. Alay, G. Akar","doi":"10.1109/SIU.2006.1659793","DOIUrl":"https://doi.org/10.1109/SIU.2006.1659793","url":null,"abstract":"In H.264 video coding standard, in order to increase the coding efficiency, several new techniques such as spatial intra prediction, integer transformation have been introduced. However these also increase the computational complexity drastically. In this paper, we propose a new fast intra prediction algorithm to be used instead of the full search algorithm to choose the best mode for spatial intra prediction. Experimental results show that, our algorithm achieves 52% computation reduction on the average while maintaining similar PSNR and an average bitrate increase of 1.5%","PeriodicalId":415037,"journal":{"name":"2006 IEEE 14th Signal Processing and Communications Applications","volume":"53 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2006-04-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132708851","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：