Pub Date : 2006-04-17DOI: 10.1109/SIU.2006.1659738
M. Dogan, O. Gerek
In this paper, orthogonality of block wavelet transforms (BWT) is shown. First the orthogonality of one-stage and two-stage block wavelet transforms is shown, then the result is generalized for multi-stage block wavelet transform
研究了块小波变换的正交性。首先给出了一级和二级分块小波变换的正交性，然后将其推广到多级分块小波变换
{"title":"On the Orthogonality of Block Wavelet Transforms","authors":"M. Dogan, O. Gerek","doi":"10.1109/SIU.2006.1659738","DOIUrl":"https://doi.org/10.1109/SIU.2006.1659738","url":null,"abstract":"In this paper, orthogonality of block wavelet transforms (BWT) is shown. First the orthogonality of one-stage and two-stage block wavelet transforms is shown, then the result is generalized for multi-stage block wavelet transform","PeriodicalId":415037,"journal":{"name":"2006 IEEE 14th Signal Processing and Communications Applications","volume":"3 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":"125654995","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.2006.1659696
M. Karabatak, A. Şengur, M. C. Ince
Texture can be defined as a local statistical pattern of texture primitives in observer's domain of interest. Texture classification aims to assign texture labels to unknown textures, according to training samples and classification rules. Association rules capture both structural and statistical information, and automatically identify the structures that occur most frequently and relationships that have significant discriminative power. So, association rules can be adapted to capture frequently occurring local structures in textures. This paper describes the usage of association rules for texture classification problem. The performed experimental studies show the effectiveness of the association rules
{"title":"Texture Classification Using Edge Detection and Association Rules","authors":"M. Karabatak, A. Şengur, M. C. Ince","doi":"10.1109/SIU.2006.1659696","DOIUrl":"https://doi.org/10.1109/SIU.2006.1659696","url":null,"abstract":"Texture can be defined as a local statistical pattern of texture primitives in observer's domain of interest. Texture classification aims to assign texture labels to unknown textures, according to training samples and classification rules. Association rules capture both structural and statistical information, and automatically identify the structures that occur most frequently and relationships that have significant discriminative power. So, association rules can be adapted to capture frequently occurring local structures in textures. This paper describes the usage of association rules for texture classification problem. The performed experimental studies show the effectiveness of the association rules","PeriodicalId":415037,"journal":{"name":"2006 IEEE 14th Signal Processing and Communications Applications","volume":"14 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":"123720007","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}
{"title":"Boosting Classifiers for Music Genre Classification","authors":"Ulas Bagci, E. Erzin","doi":"10.1007/11569596_60","DOIUrl":"https://doi.org/10.1007/11569596_60","url":null,"abstract":"","PeriodicalId":415037,"journal":{"name":"2006 IEEE 14th Signal Processing and Communications Applications","volume":"34 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":"132410032","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.2006.1659860
N. Ozbek, A. Tekalp
Long-term reference prediction is an important feature of the H.264/MPEG-4 AVC standard, which provides a tradeoff between compression gain and computational complexity. In this study, we propose a long-term reference selection method for videos with frequent camera transitions to optimize compression efficiency at shot boundaries without increasing the computational complexity. Experimental results show up to 50% reduction in the number of bits (at the same PSNR) for frames at the border of camera transitions
{"title":"H.264 Long-Term Reference Selection for Videos with Frequent Camera Transitions","authors":"N. Ozbek, A. Tekalp","doi":"10.1109/SIU.2006.1659860","DOIUrl":"https://doi.org/10.1109/SIU.2006.1659860","url":null,"abstract":"Long-term reference prediction is an important feature of the H.264/MPEG-4 AVC standard, which provides a tradeoff between compression gain and computational complexity. In this study, we propose a long-term reference selection method for videos with frequent camera transitions to optimize compression efficiency at shot boundaries without increasing the computational complexity. Experimental results show up to 50% reduction in the number of bits (at the same PSNR) for frames at the border of camera transitions","PeriodicalId":415037,"journal":{"name":"2006 IEEE 14th Signal Processing and Communications Applications","volume":"46 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":"132851723","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.2006.1659820
O. Aran
In this study, using a two camera setup, we designed a system that recognizes 3D gestures. When 3D reconstruction is not possible or infeasible, combining 2D hand trajectories at feature or decision level increases the system performance drastically. The trajectories are extracted by tracking the center-of-mass of the hand and the width, height and orientation of the enclosing ellipse. Trajectories are then smoothed using a Kalman filter. Following the translation and scale normalization, the trajectories are modelled using hidden Markov models (HMM) and using support vector machines (SVM) by converting the trajectories to fixed length using re-sampling. Trajectories extracted from different cameras are combined at different levels and the effect to the system performance is observed. The best result is obtained by modelling the trajectories using HMMs and combining at decision level, with %1 error in 210 test examples
{"title":"Combination Strategies for 2D Features to Recognize 3D Gestures","authors":"O. Aran","doi":"10.1109/SIU.2006.1659820","DOIUrl":"https://doi.org/10.1109/SIU.2006.1659820","url":null,"abstract":"In this study, using a two camera setup, we designed a system that recognizes 3D gestures. When 3D reconstruction is not possible or infeasible, combining 2D hand trajectories at feature or decision level increases the system performance drastically. The trajectories are extracted by tracking the center-of-mass of the hand and the width, height and orientation of the enclosing ellipse. Trajectories are then smoothed using a Kalman filter. Following the translation and scale normalization, the trajectories are modelled using hidden Markov models (HMM) and using support vector machines (SVM) by converting the trajectories to fixed length using re-sampling. Trajectories extracted from different cameras are combined at different levels and the effect to the system performance is observed. The best result is obtained by modelling the trajectories using HMMs and combining at decision level, with %1 error in 210 test examples","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":"130800542","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.2006.1659786
K. Gunel, R. Asliyan
In this study, we have designed and implemented a system, which uses n-gram statistical language model in order to facilitate optical character recognition, speech synthesis and recognition systems. First, the syllables bigram frequencies are extracted from Turkish corpora. Then, the test database including the words, which are written correctly and wrongly, is created. The probability of the words appears the given text is calculated and the wrongly and, correctly written words are determined. The system finds the wrongly written words about 86.13% with the proposed approach and the correctly written words are found about 88.32%
{"title":"Turkish Word Error Detection Using Syllable Bigram Statistics","authors":"K. Gunel, R. Asliyan","doi":"10.1109/SIU.2006.1659786","DOIUrl":"https://doi.org/10.1109/SIU.2006.1659786","url":null,"abstract":"In this study, we have designed and implemented a system, which uses n-gram statistical language model in order to facilitate optical character recognition, speech synthesis and recognition systems. First, the syllables bigram frequencies are extracted from Turkish corpora. Then, the test database including the words, which are written correctly and wrongly, is created. The probability of the words appears the given text is calculated and the wrongly and, correctly written words are determined. The system finds the wrongly written words about 86.13% with the proposed approach and the correctly written words are found about 88.32%","PeriodicalId":415037,"journal":{"name":"2006 IEEE 14th Signal Processing and Communications Applications","volume":"62 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":"127939821","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.2006.1659740
M. Torun, A. Ozkurt
This study offers an alternative method for speed detection of the automobiles on a highway. The system we develop, captures live video from two cameras, meaning left and right, simultaneously and obtains the difference of the successive frames. Then, it converts the difference frame into a binary image using image processing techniques and calculates the center of mass of this binary image. The distance information is calculated using stereo analysis and the difference information between mass centers of left and right frames. The speed of the moving object between t1 and t2 is calculated using the difference of calculated distance datas and Deltat=t2 -t1. The system performance is calculated as minimum 95%
{"title":"Distance and Speed Measurement using Stereo Analysis","authors":"M. Torun, A. Ozkurt","doi":"10.1109/SIU.2006.1659740","DOIUrl":"https://doi.org/10.1109/SIU.2006.1659740","url":null,"abstract":"This study offers an alternative method for speed detection of the automobiles on a highway. The system we develop, captures live video from two cameras, meaning left and right, simultaneously and obtains the difference of the successive frames. Then, it converts the difference frame into a binary image using image processing techniques and calculates the center of mass of this binary image. The distance information is calculated using stereo analysis and the difference information between mass centers of left and right frames. The speed of the moving object between t1 and t2 is calculated using the difference of calculated distance datas and Deltat=t2 -t1. The system performance is calculated as minimum 95%","PeriodicalId":415037,"journal":{"name":"2006 IEEE 14th Signal Processing and Communications Applications","volume":"142 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":"121520634","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.2006.1659878
Y. Altug, N. P. Ayerden, I. Erguler, E. Anarim
In 2004, A. Gouget and H. Sibert proposed a new keystream generator called the bit search generator (BSG), to provide high resistance against algebraic attacks. BSG has a very simple algorithm and attractive properties. However it has been cryptanalyzed in different studies by using the fact that output of BSG can be uniquely expressed by differential of the input sequence. Recently, Gouget et al. introduced two modified versions of BSG, named as MBSG and ABSG, to increase its security and also presented their security analysis in the same paper. The best attack that they give against ABSG and MBSG has complexity O(2L/2) and requires O(L2L/2) bits of keystream. In this study, we have shown that BSG, MBSG and ABSG can be cryptanalyzed with a time complexity O(2L/3) by using a time-memory trade-off attack. The method requires 22L/3 words of memory and O(2L+2L/3) bits of keystream. According to computer simulation results, we have found out that MBSG is the most vulnerable generator among BSG and variants to proposed attack. Moreover, ABSG doesn't bring any additional security to original BSG for proposed time-memory trade-off attack
2004年，a . Gouget和H. Sibert提出了一种新的密钥流生成器，称为比特搜索生成器(BSG)，以提供高抗代数攻击的能力。BSG具有非常简单的算法和吸引人的特性。然而，在不同的研究中，利用BSG的输出可以由输入序列的微分唯一表示这一事实对其进行了密码分析。最近，Gouget et al.引入了BSG的两个修改版本，分别命名为MBSG和ABSG，以增加其安全性，并在同一篇论文中介绍了他们的安全性分析。他们给出的针对ABSG和MBSG的最佳攻击复杂度为0 (2L/2)，并且需要O(L2L/2)位密钥流。在这项研究中，我们已经证明了BSG, MBSG和ABSG可以使用时间-记忆权衡攻击以时间复杂度0 (2L/3)进行密码分析。该方法需要22L/3个字的内存和O(2L+2L/3)位的密钥流。根据计算机仿真结果，我们发现在BSG和变体中，MBSG是最容易受到攻击的发生器。此外，对于所提出的时间-内存权衡攻击，ABSG不会给原始BSG带来任何额外的安全性
{"title":"A Time-Memory Trade-off Attack to Bit Search Generator and Its Variants","authors":"Y. Altug, N. P. Ayerden, I. Erguler, E. Anarim","doi":"10.1109/SIU.2006.1659878","DOIUrl":"https://doi.org/10.1109/SIU.2006.1659878","url":null,"abstract":"In 2004, A. Gouget and H. Sibert proposed a new keystream generator called the bit search generator (BSG), to provide high resistance against algebraic attacks. BSG has a very simple algorithm and attractive properties. However it has been cryptanalyzed in different studies by using the fact that output of BSG can be uniquely expressed by differential of the input sequence. Recently, Gouget et al. introduced two modified versions of BSG, named as MBSG and ABSG, to increase its security and also presented their security analysis in the same paper. The best attack that they give against ABSG and MBSG has complexity O(2L/2) and requires O(L2L/2) bits of keystream. In this study, we have shown that BSG, MBSG and ABSG can be cryptanalyzed with a time complexity O(2L/3) by using a time-memory trade-off attack. The method requires 22L/3 words of memory and O(2L+2L/3) bits of keystream. According to computer simulation results, we have found out that MBSG is the most vulnerable generator among BSG and variants to proposed attack. Moreover, ABSG doesn't bring any additional security to original BSG for proposed time-memory trade-off attack","PeriodicalId":415037,"journal":{"name":"2006 IEEE 14th Signal Processing and Communications Applications","volume":"2011 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":"121623505","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.2006.1659814
H. Ozkaramali, A. Baradarani, H. Demirel, B. Ozmen, T. Çelik
Moving object edge detection and segmentation method is presented with utilizing multi-wavelets. The subsequent segmentation of moving objects is achieved by binary morphological operations. The proposed multi-wavelet based method is compared with the methods based on scalar wavelets using both single and double change detection techniques. The simulation results indicate that multi-wavelets with repeated row pre-processing employing double change detection method outperform scalar wavelet-based methods in the number of detected moving edges and better preserve the moving edges. As a result the quality of moving object segmentation has been improved over the scalar methods
{"title":"Moving Object Edge Detection and Segmentation using Multi-Wavelets","authors":"H. Ozkaramali, A. Baradarani, H. Demirel, B. Ozmen, T. Çelik","doi":"10.1109/SIU.2006.1659814","DOIUrl":"https://doi.org/10.1109/SIU.2006.1659814","url":null,"abstract":"Moving object edge detection and segmentation method is presented with utilizing multi-wavelets. The subsequent segmentation of moving objects is achieved by binary morphological operations. The proposed multi-wavelet based method is compared with the methods based on scalar wavelets using both single and double change detection techniques. The simulation results indicate that multi-wavelets with repeated row pre-processing employing double change detection method outperform scalar wavelet-based methods in the number of detected moving edges and better preserve the moving edges. As a result the quality of moving object segmentation has been improved over the scalar methods","PeriodicalId":415037,"journal":{"name":"2006 IEEE 14th Signal Processing and Communications Applications","volume":"69 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":"124920242","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.2006.1659697
A. Kanak, Y. Bicil, M. U. Dogan, H. Palaz
Recognition engines are common tools for both speech and speaker recognition. With this respect, TREN-SI (Turkish recognition engine for speaker identification) is presented as a hidden Markov model-based (HMM-based), two-layered distributed speaker identification software. TREN-SI contains specialized modules that allow a full interoperable platform including a speaker recognizer, feature extractor and a performance monitoring module. TREN-SI has basically two layers: First layer is the central server that distributes the calls acquired from different people to the appropriate remote servers according to their current CPU load of the recognition process after some speech signal preprocessing and the second layer consists of the remote servers which performs the critical speaker recognition task. This component-based architecture enables TREN-SI applicable to distributed environments. TREN-SI is developed as a solution especially for physical or logical access control problems considering user authentication and authorization
{"title":"TREN-SI: A DCOM-Based Speaker Identification Software","authors":"A. Kanak, Y. Bicil, M. U. Dogan, H. Palaz","doi":"10.1109/SIU.2006.1659697","DOIUrl":"https://doi.org/10.1109/SIU.2006.1659697","url":null,"abstract":"Recognition engines are common tools for both speech and speaker recognition. With this respect, TREN-SI (Turkish recognition engine for speaker identification) is presented as a hidden Markov model-based (HMM-based), two-layered distributed speaker identification software. TREN-SI contains specialized modules that allow a full interoperable platform including a speaker recognizer, feature extractor and a performance monitoring module. TREN-SI has basically two layers: First layer is the central server that distributes the calls acquired from different people to the appropriate remote servers according to their current CPU load of the recognition process after some speech signal preprocessing and the second layer consists of the remote servers which performs the critical speaker recognition task. This component-based architecture enables TREN-SI applicable to distributed environments. TREN-SI is developed as a solution especially for physical or logical access control problems considering user authentication and authorization","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":"125199378","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：