Pub Date : 1990-12-01DOI: 10.1109/ICASSP.1987.1169580
Douglas L. Jones, T. Parks
We present a data-adaptive time-frequency representation that obtains high resolution of signal components in time-frequency. This representation overcomes the often poor resolution of the traditional short-time Fourier transform, while avoiding the nonlinearities that make the Wigner distribution and other bilinear representations difficult to interpret and use. The new method uses adaptive Gaussian windows, with the window parameters varying at different time-frequency locations to maximize the local signal concentration in time-frequency. Two methods for selecting the Gaussian parameters are presented: a parameter estimation approach, and a method that maximizes a measure of local signal concentration.
{"title":"A high resolution data-adaptive time-frequency representation","authors":"Douglas L. Jones, T. Parks","doi":"10.1109/ICASSP.1987.1169580","DOIUrl":"https://doi.org/10.1109/ICASSP.1987.1169580","url":null,"abstract":"We present a data-adaptive time-frequency representation that obtains high resolution of signal components in time-frequency. This representation overcomes the often poor resolution of the traditional short-time Fourier transform, while avoiding the nonlinearities that make the Wigner distribution and other bilinear representations difficult to interpret and use. The new method uses adaptive Gaussian windows, with the window parameters varying at different time-frequency locations to maximize the local signal concentration in time-frequency. Two methods for selecting the Gaussian parameters are presented: a parameter estimation approach, and a method that maximizes a measure of local signal concentration.","PeriodicalId":140810,"journal":{"name":"ICASSP '87. IEEE International Conference on Acoustics, Speech, and Signal Processing","volume":"73 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1990-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132658656","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 : 1989-05-01DOI: 10.1109/ICASSP.1987.1169788
G. Giannakis, J. Mendel, Xiaofeng Zhao
Based on the Maximum-Likelihood principle, we develop a locally optimal method for detecting the location and estimating the amplitude of spikes in a sequence, which are considered the random input of a known ARMA model. A Bernoulli-Gaussian product model is adopted for the sparse-spike sequence, and the available data consist of a single, noisy, output record. By employing a Prediction-Error formulation our iterative algorithm guarantees the increase of a unique likelihood function used for the combined estimation/detection problem. Amplitude estimation is carried out with Kalman smoothing techniques, and event detection is performed in two ways, as an event adder and as an event remover. Synthetic examples verify that our algorithm is self-initialized, consistent, and fast.
{"title":"A fast prediction-error detector for estimating sparse-spike sequences","authors":"G. Giannakis, J. Mendel, Xiaofeng Zhao","doi":"10.1109/ICASSP.1987.1169788","DOIUrl":"https://doi.org/10.1109/ICASSP.1987.1169788","url":null,"abstract":"Based on the Maximum-Likelihood principle, we develop a locally optimal method for detecting the location and estimating the amplitude of spikes in a sequence, which are considered the random input of a known ARMA model. A Bernoulli-Gaussian product model is adopted for the sparse-spike sequence, and the available data consist of a single, noisy, output record. By employing a Prediction-Error formulation our iterative algorithm guarantees the increase of a unique likelihood function used for the combined estimation/detection problem. Amplitude estimation is carried out with Kalman smoothing techniques, and event detection is performed in two ways, as an event adder and as an event remover. Synthetic examples verify that our algorithm is self-initialized, consistent, and fast.","PeriodicalId":140810,"journal":{"name":"ICASSP '87. IEEE International Conference on Acoustics, Speech, and Signal Processing","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1989-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114234610","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 : 1987-12-23DOI: 10.1109/ICASSP.1987.1169668
T. R. Esselman, J. Verly
Although little known, mathematical morphology (MM) offers great potential in the areas of image enhancement, feature extraction, and object recognition. MM has the intrinsic ability to quantitatively analyze object shapes in both 2 and 3 dimensions. Using MM to extract features and recognize objects in range imagery seems particularly appropriate since range data is a natural source of shape information. We present several experimental results of applying MM techniques to real and synthetic range imagery, both for noise removal and feature extraction.
{"title":"Some applications of mathematical morphology to range imagery","authors":"T. R. Esselman, J. Verly","doi":"10.1109/ICASSP.1987.1169668","DOIUrl":"https://doi.org/10.1109/ICASSP.1987.1169668","url":null,"abstract":"Although little known, mathematical morphology (MM) offers great potential in the areas of image enhancement, feature extraction, and object recognition. MM has the intrinsic ability to quantitatively analyze object shapes in both 2 and 3 dimensions. Using MM to extract features and recognize objects in range imagery seems particularly appropriate since range data is a natural source of shape information. We present several experimental results of applying MM techniques to real and synthetic range imagery, both for noise removal and feature extraction.","PeriodicalId":140810,"journal":{"name":"ICASSP '87. IEEE International Conference on Acoustics, Speech, and Signal Processing","volume":"2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1987-12-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126562305","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 : 1987-04-06DOI: 10.1109/ICASSP.1987.1169328
Lei Xu, P. Yan, Tong Chang
In this paper, the reconstruction of discrete signal with finite time duration from its end point and Fourier Transform (FT) magnitude is considered. Based on one result of [1] that a class of discrete signal can be reconstructed from its FT magnitude and one end sample point, with the help of Measure Theory, furtherly we point out that a correspondence between RN+1space and discrete signals with duration of N+1 points can be set up, and the signals that can't be reconstructed from its end point and FT magnitude correspond to a subset of RN+1with measure zero. In other words, discrete signal with finite time duration can almost be uniquely reconstructed from its end point and FT magnitude.
{"title":"Almost unique specification of discrete finite length signal: From its end point and Fourier transform magnitude","authors":"Lei Xu, P. Yan, Tong Chang","doi":"10.1109/ICASSP.1987.1169328","DOIUrl":"https://doi.org/10.1109/ICASSP.1987.1169328","url":null,"abstract":"In this paper, the reconstruction of discrete signal with finite time duration from its end point and Fourier Transform (FT) magnitude is considered. Based on one result of [1] that a class of discrete signal can be reconstructed from its FT magnitude and one end sample point, with the help of Measure Theory, furtherly we point out that a correspondence between RN+1space and discrete signals with duration of N+1 points can be set up, and the signals that can't be reconstructed from its end point and FT magnitude correspond to a subset of RN+1with measure zero. In other words, discrete signal with finite time duration can almost be uniquely reconstructed from its end point and FT magnitude.","PeriodicalId":140810,"journal":{"name":"ICASSP '87. IEEE International Conference on Acoustics, Speech, and Signal Processing","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1987-04-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126196346","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 : 1987-04-06DOI: 10.1109/ICASSP.1987.1169901
M. Manry, C. T. Huddleston
Optimal parameter estimation algorithms are developed using the maximum likelihood technique, when no statistics are available for the parameter. Sub-optimal parameter estimates, using one sample of the autocorrelation of the DFT, have been developed previously. In this paper, maximum likelihood estimates are derived, given the auto-correlation function of the received signal's DFT. These estimates sometimes require less computation time than conventional estimates, and frequently have a closed form or simple iterative implementation.
{"title":"Parameter estimation using the autocorrelation of the discrete Fourier transform","authors":"M. Manry, C. T. Huddleston","doi":"10.1109/ICASSP.1987.1169901","DOIUrl":"https://doi.org/10.1109/ICASSP.1987.1169901","url":null,"abstract":"Optimal parameter estimation algorithms are developed using the maximum likelihood technique, when no statistics are available for the parameter. Sub-optimal parameter estimates, using one sample of the autocorrelation of the DFT, have been developed previously. In this paper, maximum likelihood estimates are derived, given the auto-correlation function of the received signal's DFT. These estimates sometimes require less computation time than conventional estimates, and frequently have a closed form or simple iterative implementation.","PeriodicalId":140810,"journal":{"name":"ICASSP '87. IEEE International Conference on Acoustics, Speech, and Signal Processing","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1987-04-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114177828","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 : 1987-04-06DOI: 10.1109/ICASSP.1987.1169578
Vishwa Gupta, Matthew Lennig, P. Mermelstein
This paper proposes a new way of using vector quantization for improving recognition performance for a 60,000 word vocabulary speaker-trained isolated word recognizer using a phonemic Markov model approach to speech recognition. We show that we can effectively increase the codebook size by dividing the feature vector into two vectors of lower dimensionality, and then quantizing and training each vector separately. For a small codebook size, integration of the results of the two parameter vectors provides significant improvement in recognition performance as compared to the quantizing and training of the entire feature set together. Even for a codebook size as small as 64, the results obtained when using the new quantization procedure are quite close to those obtained when using Gaussian distribution of the parameter vectors.
{"title":"Integration of acoustic information in a large vocabulary word recognizer","authors":"Vishwa Gupta, Matthew Lennig, P. Mermelstein","doi":"10.1109/ICASSP.1987.1169578","DOIUrl":"https://doi.org/10.1109/ICASSP.1987.1169578","url":null,"abstract":"This paper proposes a new way of using vector quantization for improving recognition performance for a 60,000 word vocabulary speaker-trained isolated word recognizer using a phonemic Markov model approach to speech recognition. We show that we can effectively increase the codebook size by dividing the feature vector into two vectors of lower dimensionality, and then quantizing and training each vector separately. For a small codebook size, integration of the results of the two parameter vectors provides significant improvement in recognition performance as compared to the quantizing and training of the entire feature set together. Even for a codebook size as small as 64, the results obtained when using the new quantization procedure are quite close to those obtained when using Gaussian distribution of the parameter vectors.","PeriodicalId":140810,"journal":{"name":"ICASSP '87. IEEE International Conference on Acoustics, Speech, and Signal Processing","volume":"94 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1987-04-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128707518","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 : 1987-04-06DOI: 10.1109/ICASSP.1987.1169380
A. Lowry, Sqama Hossain, W. Millar
This paper presents a data structure based on the k-d binary tree which substantially reduces the search complexity of a full search vector quantiser with negligible degradation in signal-to-noise ratio. The search complexity isk + O(logN)rather than N for a codebook of dimension k and size N. Special features of the structure are (1) the use of a rotational transform prior to encoding and (2) the computational efficiency of the design algorithm due to the simple structure of the k-d tree.
{"title":"Binary search trees for vector quantisation","authors":"A. Lowry, Sqama Hossain, W. Millar","doi":"10.1109/ICASSP.1987.1169380","DOIUrl":"https://doi.org/10.1109/ICASSP.1987.1169380","url":null,"abstract":"This paper presents a data structure based on the k-d binary tree which substantially reduces the search complexity of a full search vector quantiser with negligible degradation in signal-to-noise ratio. The search complexity isk + O(logN)rather than N for a codebook of dimension k and size N. Special features of the structure are (1) the use of a rotational transform prior to encoding and (2) the computational efficiency of the design algorithm due to the simple structure of the k-d tree.","PeriodicalId":140810,"journal":{"name":"ICASSP '87. IEEE International Conference on Acoustics, Speech, and Signal Processing","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1987-04-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128733587","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 : 1987-04-06DOI: 10.1109/ICASSP.1987.1169598
S. Kimura, Y. Nara
This paper describes an interactive extraction of phonemic variation rules in continuous speech spoken by multiple speakers. To realize a continuous speech recognizer, we must first develop a highly accurate phoneme recognizer. The major problem related to phoneme recognizers is the phonemic variations in continuous speech. Our work focuses on the interactive analysis of phonemic variations in continuous speech and the extraction of the phonemic variation rules for many speakers. We extracted 317 rules related to 21 kinds of phonemic variation phenomena from 10,000 Japanese-language phrases spoken by 10 male speakers. With these rules, 97.6% of 36,000 Japanese-language phrases spoken by 36 test speakers (30 males and 6 females) were correctly segmented by our top-down phoneme segmentation system. Furthermore, a subset of the rules for each speaker was automatically obtained. On average, each subset contains 53.2% of the rules.
{"title":"Extraction of phonemic variation rules in continuous speech spoken by multiple speakers","authors":"S. Kimura, Y. Nara","doi":"10.1109/ICASSP.1987.1169598","DOIUrl":"https://doi.org/10.1109/ICASSP.1987.1169598","url":null,"abstract":"This paper describes an interactive extraction of phonemic variation rules in continuous speech spoken by multiple speakers. To realize a continuous speech recognizer, we must first develop a highly accurate phoneme recognizer. The major problem related to phoneme recognizers is the phonemic variations in continuous speech. Our work focuses on the interactive analysis of phonemic variations in continuous speech and the extraction of the phonemic variation rules for many speakers. We extracted 317 rules related to 21 kinds of phonemic variation phenomena from 10,000 Japanese-language phrases spoken by 10 male speakers. With these rules, 97.6% of 36,000 Japanese-language phrases spoken by 36 test speakers (30 males and 6 females) were correctly segmented by our top-down phoneme segmentation system. Furthermore, a subset of the rules for each speaker was automatically obtained. On average, each subset contains 53.2% of the rules.","PeriodicalId":140810,"journal":{"name":"ICASSP '87. IEEE International Conference on Acoustics, Speech, and Signal Processing","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1987-04-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128980028","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 : 1987-04-06DOI: 10.1109/ICASSP.1987.1169423
M. Al-Kindi, J. Dunlop
This paper describes an adaptive noise cancelling structure suitable for situations where the noise reference transducer is closely spaced relative to the primary transducer. The structure is based on two LMS delay line cancellers with cross coupled feedback. This structure is shown, under certain circumstances, to cancel noise with low signal distortion when the transmission paths between primary and secondary sensors have low attenuation and the primary signal is continuous. The system is shown to have an enhanced performance when the primary signal is intermittent and a signal energy detector is used.
{"title":"A low distortion adaptive noise cancellation structure for real time applications","authors":"M. Al-Kindi, J. Dunlop","doi":"10.1109/ICASSP.1987.1169423","DOIUrl":"https://doi.org/10.1109/ICASSP.1987.1169423","url":null,"abstract":"This paper describes an adaptive noise cancelling structure suitable for situations where the noise reference transducer is closely spaced relative to the primary transducer. The structure is based on two LMS delay line cancellers with cross coupled feedback. This structure is shown, under certain circumstances, to cancel noise with low signal distortion when the transmission paths between primary and secondary sensors have low attenuation and the primary signal is continuous. The system is shown to have an enhanced performance when the primary signal is intermittent and a signal energy detector is used.","PeriodicalId":140810,"journal":{"name":"ICASSP '87. IEEE International Conference on Acoustics, Speech, and Signal Processing","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1987-04-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129161302","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 : 1987-04-06DOI: 10.1109/ICASSP.1987.1169665
E. B. Hinkle, J. Sanz
This paper describes the use of an image contrast measure for producing binary segmentations of images in a certain class of applications. This method is well-suited for fast pipeline implementations, because the contrast measure uses only two local features in the image. To eliminate segmentation noise, we post-process the segmentations using binary morphological operations. This method has been applied to three different microelectronics inspection problems, with consistently good results, and experimental results from each of these applications are presented here. Also, we discuss this technique in terms of the theory of polynomial classifiers.
{"title":"Fast image segmentation for some machine vision applications","authors":"E. B. Hinkle, J. Sanz","doi":"10.1109/ICASSP.1987.1169665","DOIUrl":"https://doi.org/10.1109/ICASSP.1987.1169665","url":null,"abstract":"This paper describes the use of an image contrast measure for producing binary segmentations of images in a certain class of applications. This method is well-suited for fast pipeline implementations, because the contrast measure uses only two local features in the image. To eliminate segmentation noise, we post-process the segmentations using binary morphological operations. This method has been applied to three different microelectronics inspection problems, with consistently good results, and experimental results from each of these applications are presented here. Also, we discuss this technique in terms of the theory of polynomial classifiers.","PeriodicalId":140810,"journal":{"name":"ICASSP '87. IEEE International Conference on Acoustics, Speech, and Signal Processing","volume":"85 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"1987-04-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114735010","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：