Pub Date : 2004-03-28DOI: 10.1109/IAI.2004.1300933
M. Ponti, N. Mascarenhas
Methods for material analysis on images are essential in many applications. We present a set of experiments with classifier combiners in order to recognize materials in multispectral images with applications in soil science. These images were obtained by the transmission of different energies with a computerized tomography scanner. The use of the linear attenuation coefficients as classification features is studied. The multispectral images are classified, and combining techniques for the classifiers are investigated. A comparison of the combiners with the individual classifiers is also performed.
{"title":"Material analysis on noisy multispectral images using classifier combination","authors":"M. Ponti, N. Mascarenhas","doi":"10.1109/IAI.2004.1300933","DOIUrl":"https://doi.org/10.1109/IAI.2004.1300933","url":null,"abstract":"Methods for material analysis on images are essential in many applications. We present a set of experiments with classifier combiners in order to recognize materials in multispectral images with applications in soil science. These images were obtained by the transmission of different energies with a computerized tomography scanner. The use of the linear attenuation coefficients as classification features is studied. The multispectral images are classified, and combining techniques for the classifiers are investigated. A comparison of the combiners with the individual classifiers is also performed.","PeriodicalId":326040,"journal":{"name":"6th IEEE Southwest Symposium on Image Analysis and Interpretation, 2004.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2004-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115304452","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 : 2004-03-28DOI: 10.1109/IAI.2004.1300968
P. A. Mlsna, N. Sirakov
We describe the development of novel and efficient approaches and algorithms for a medical image content-based retrieval system capable of extracting and indexing key information about region shape. First, the general structure and the main components of the system are discussed. For grayscale segmentation to locate regions, we have explored a fast active contour approach based on the geometric heat differential equation. Region representation involves a set of extracted shape-based features. A technique for feature organization using N-dimensional feature vectors is employed. The image retrieval process compares similarity of query vectors to the indexed feature vectors. A convex hull model using the heat differential equation is used to organize the index of features to reduce the search space. Some experiments have been performed to test and validate certain portions of our approach. Finally; advantages and disadvantages together with the computational complexity of this system are discussed.
{"title":"Intelligent shape feature extraction and indexing for efficient content-based medical image retrieval","authors":"P. A. Mlsna, N. Sirakov","doi":"10.1109/IAI.2004.1300968","DOIUrl":"https://doi.org/10.1109/IAI.2004.1300968","url":null,"abstract":"We describe the development of novel and efficient approaches and algorithms for a medical image content-based retrieval system capable of extracting and indexing key information about region shape. First, the general structure and the main components of the system are discussed. For grayscale segmentation to locate regions, we have explored a fast active contour approach based on the geometric heat differential equation. Region representation involves a set of extracted shape-based features. A technique for feature organization using N-dimensional feature vectors is employed. The image retrieval process compares similarity of query vectors to the indexed feature vectors. A convex hull model using the heat differential equation is used to organize the index of features to reduce the search space. Some experiments have been performed to test and validate certain portions of our approach. Finally; advantages and disadvantages together with the computational complexity of this system are discussed.","PeriodicalId":326040,"journal":{"name":"6th IEEE Southwest Symposium on Image Analysis and Interpretation, 2004.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2004-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127201799","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 : 2004-03-28DOI: 10.1109/IAI.2004.1300967
P. C. Tay, J. Havlicek
The paper implements the discrete wavelet transform in the discrete Fourier domain. The need for such an approach arose out of our desire to find a convenient means of realizing a new class of non-separable orientation selective 2D wavelet filter banks that are designed directly in the DFT domain. The filter bank design process begins with a conventional separable 2D perfect reconstruction parallel filter bank that is not orientation selective. In the DFT domain, each non-low pass channel is decomposed into the sum of two orientation selective frequency responses that are each supported on only two quadrants of the 2D frequency plane. The resulting filter bank possesses the good joint localization properties of orthogonal wavelet transforms and offers both perfect reconstruction and orientation selectivity. However, the orientation selective channels are non-separable - they cannot be implemented as iterated 1D convolutions according to the usual separable 2D wavelet transform paradigm. To overcome this difficulty, we develop straightforward techniques for implementing the DWT directly in the DFT domain.
{"title":"Frequency implementation of discrete wavelet transforms","authors":"P. C. Tay, J. Havlicek","doi":"10.1109/IAI.2004.1300967","DOIUrl":"https://doi.org/10.1109/IAI.2004.1300967","url":null,"abstract":"The paper implements the discrete wavelet transform in the discrete Fourier domain. The need for such an approach arose out of our desire to find a convenient means of realizing a new class of non-separable orientation selective 2D wavelet filter banks that are designed directly in the DFT domain. The filter bank design process begins with a conventional separable 2D perfect reconstruction parallel filter bank that is not orientation selective. In the DFT domain, each non-low pass channel is decomposed into the sum of two orientation selective frequency responses that are each supported on only two quadrants of the 2D frequency plane. The resulting filter bank possesses the good joint localization properties of orthogonal wavelet transforms and offers both perfect reconstruction and orientation selectivity. However, the orientation selective channels are non-separable - they cannot be implemented as iterated 1D convolutions according to the usual separable 2D wavelet transform paradigm. To overcome this difficulty, we develop straightforward techniques for implementing the DWT directly in the DFT domain.","PeriodicalId":326040,"journal":{"name":"6th IEEE Southwest Symposium on Image Analysis and Interpretation, 2004.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2004-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114370805","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 : 2004-03-28DOI: 10.1109/IAI.2004.1300937
D.M. Thodi, J.J. Rodriguez
We propose a new reversible (lossless) watermarking algorithm for digital images. Being reversible, the algorithm enables the recovery of the original host information upon the extraction of the embedded information. The proposed technique exploits the inherent correlation among the adjacent pixels in an image region using a predictor. The information bits are embedded into the prediction errors, which enables us to embed a large payload while keeping the distortion low. A histogram shift at the encoder enables the decoder to identify the embedded location.
{"title":"Reversible watermarking by prediction-error expansion","authors":"D.M. Thodi, J.J. Rodriguez","doi":"10.1109/IAI.2004.1300937","DOIUrl":"https://doi.org/10.1109/IAI.2004.1300937","url":null,"abstract":"We propose a new reversible (lossless) watermarking algorithm for digital images. Being reversible, the algorithm enables the recovery of the original host information upon the extraction of the embedded information. The proposed technique exploits the inherent correlation among the adjacent pixels in an image region using a predictor. The information bits are embedded into the prediction errors, which enables us to embed a large payload while keeping the distortion low. A histogram shift at the encoder enables the decoder to identify the embedded location.","PeriodicalId":326040,"journal":{"name":"6th IEEE Southwest Symposium on Image Analysis and Interpretation, 2004.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2004-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129210025","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 : 2004-03-28DOI: 10.1109/IAI.2004.1300977
O. Kahler, Joachim Denzler, J. Triesch
Sensor data fusion from multiple cameras is an important problem for machine vision systems operating in complex, natural environments. We tackle the problem of how information from different sensors can be fused in 3D object tracking. We embed an approach called democratic integration into a probabilistic framework and solve the fusion step by hierarchically fusing the information of different sensors and different information sources (cues) derived from each sensor. We compare different fusion architectures and different adaptation schemes. The experiments for 3D object tracking using three calibrated cameras show that adaptive hierarchical fusion improves the tracking robustness and accuracy compared to a flat fusion strategy.
{"title":"Hierarchical sensor data fusion by probabilistic cue integration for robust 3D object tracking","authors":"O. Kahler, Joachim Denzler, J. Triesch","doi":"10.1109/IAI.2004.1300977","DOIUrl":"https://doi.org/10.1109/IAI.2004.1300977","url":null,"abstract":"Sensor data fusion from multiple cameras is an important problem for machine vision systems operating in complex, natural environments. We tackle the problem of how information from different sensors can be fused in 3D object tracking. We embed an approach called democratic integration into a probabilistic framework and solve the fusion step by hierarchically fusing the information of different sensors and different information sources (cues) derived from each sensor. We compare different fusion architectures and different adaptation schemes. The experiments for 3D object tracking using three calibrated cameras show that adaptive hierarchical fusion improves the tracking robustness and accuracy compared to a flat fusion strategy.","PeriodicalId":326040,"journal":{"name":"6th IEEE Southwest Symposium on Image Analysis and Interpretation, 2004.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2004-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130630228","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 : 2004-03-28DOI: 10.1109/IAI.2004.1300972
M. Gadkari, H. Refai, J. Sluss, T. Broughan, T. Teague, R. Naukam
The paper presents a novel method to detect and count dead and live hepatocytes (liver cells) within clusters. The new method distinguishes single cells from clusters in digital microscopic images based on cell shape rather than size. The new method shows an improvement when compared to an earlier algorithm, especially in images where cells have varying sizes.
{"title":"The detection of single hepatocytes within clusters in microscopic images","authors":"M. Gadkari, H. Refai, J. Sluss, T. Broughan, T. Teague, R. Naukam","doi":"10.1109/IAI.2004.1300972","DOIUrl":"https://doi.org/10.1109/IAI.2004.1300972","url":null,"abstract":"The paper presents a novel method to detect and count dead and live hepatocytes (liver cells) within clusters. The new method distinguishes single cells from clusters in digital microscopic images based on cell shape rather than size. The new method shows an improvement when compared to an earlier algorithm, especially in images where cells have varying sizes.","PeriodicalId":326040,"journal":{"name":"6th IEEE Southwest Symposium on Image Analysis and Interpretation, 2004.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2004-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129976564","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 : 2004-03-28DOI: 10.1109/IAI.2004.1300945
S. Pei, Jing-Ming Guo, Hua Lee
A novel progressive coding scheme is presented for the efficient display of dithered images. Dithered images are the results of thresholding original gray-level images with dithering screens. After the preprocessing of bit-interleaving, this algorithm utilizes the characteristic of the reordered image to determine the transmitting order and then progressively reconstructs the dithered image. Moreover, the dithered images are further compressed by lossy and lossless procedures. The experimental results demonstrate high-quality reconstructions while maintaining low transmitted bit rates.
{"title":"A new progressive coding algorithm of dithered images","authors":"S. Pei, Jing-Ming Guo, Hua Lee","doi":"10.1109/IAI.2004.1300945","DOIUrl":"https://doi.org/10.1109/IAI.2004.1300945","url":null,"abstract":"A novel progressive coding scheme is presented for the efficient display of dithered images. Dithered images are the results of thresholding original gray-level images with dithering screens. After the preprocessing of bit-interleaving, this algorithm utilizes the characteristic of the reordered image to determine the transmitting order and then progressively reconstructs the dithered image. Moreover, the dithered images are further compressed by lossy and lossless procedures. The experimental results demonstrate high-quality reconstructions while maintaining low transmitted bit rates.","PeriodicalId":326040,"journal":{"name":"6th IEEE Southwest Symposium on Image Analysis and Interpretation, 2004.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2004-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130734836","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 : 2004-03-28DOI: 10.1109/IAI.2004.1300970
Benjamin M. Howe, A. Gururajan, H. Sari-Sarraf, L. R. Long
The paper describes a semi-automatic segmentation method for application to cervical and lumbar X-ray images. The method consists of a three stage, coarse to fine, segmentation process utilizing the generalised Hough transform for one stage, and active appearance models for two stages. Customizations to these algorithms are introduced, and segmentation results for 273 cervical X-ray images and 262 lumbar X-ray images are presented.
{"title":"Hierarchical segmentation of cervical and lumbar vertebrae using a customized generalized Hough transform and extensions to active appearance models","authors":"Benjamin M. Howe, A. Gururajan, H. Sari-Sarraf, L. R. Long","doi":"10.1109/IAI.2004.1300970","DOIUrl":"https://doi.org/10.1109/IAI.2004.1300970","url":null,"abstract":"The paper describes a semi-automatic segmentation method for application to cervical and lumbar X-ray images. The method consists of a three stage, coarse to fine, segmentation process utilizing the generalised Hough transform for one stage, and active appearance models for two stages. Customizations to these algorithms are introduced, and segmentation results for 273 cervical X-ray images and 262 lumbar X-ray images are presented.","PeriodicalId":326040,"journal":{"name":"6th IEEE Southwest Symposium on Image Analysis and Interpretation, 2004.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2004-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131687934","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 : 2004-03-28DOI: 10.1109/IAI.2004.1300956
F. Brisc
The reconstruction of 3D models from camera views is becoming an element of major importance in many applications that simulate physical interaction with the real world. 3D models are amenable to access through diverse representation modalities that typically imply trade-offs between level of detail, interaction, and computational costs. We present a 3D volumetric reconstruction method that allows users to control selectively the complexity of different surface regions, while requiring only simple 2D image editing operations. An initial reconstruction at coarse resolution is followed by an iterative refining of the surface areas corresponding to the selected regions.
{"title":"Multi-resolution volumetric reconstruction using labeled regions","authors":"F. Brisc","doi":"10.1109/IAI.2004.1300956","DOIUrl":"https://doi.org/10.1109/IAI.2004.1300956","url":null,"abstract":"The reconstruction of 3D models from camera views is becoming an element of major importance in many applications that simulate physical interaction with the real world. 3D models are amenable to access through diverse representation modalities that typically imply trade-offs between level of detail, interaction, and computational costs. We present a 3D volumetric reconstruction method that allows users to control selectively the complexity of different surface regions, while requiring only simple 2D image editing operations. An initial reconstruction at coarse resolution is followed by an iterative refining of the surface areas corresponding to the selected regions.","PeriodicalId":326040,"journal":{"name":"6th IEEE Southwest Symposium on Image Analysis and Interpretation, 2004.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2004-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114306581","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 : 2004-03-28DOI: 10.1109/IAI.2004.1300978
P. Aeluri, V. Bojan, S. Richie, A. Weeks
Video quality estimation methods in the literature can be broadly classified into two classes: objective and subjective. The metrics obtained from both classes of methods represent different intrinsic features of a video (Eude, T. and Mayache, A., Proc. ICSP'98, 1998). However, we have performed an objective quality evaluation on various videos, which range from no motion to high motion, encoded by three encoders (MPEG-1, MPEG-2 and Windows Media formats) at two different frame rates, two frame sizes and three low bit rate combinations for the purpose of determining the change in video content introduced by the encoding and decoding process. The effect of the encoding, bit rates, frame rates and frame sizes on the quality of videos has been studied. We have defined and used a quality metric, cumulative brightness error. It is the cumulative mean square error of each frame's color components, averaged for the entire video sequence. It enables comparison of video formats based on content differences and encoding parameter values.
文献中的视频质量估计方法大致可以分为两类:客观的和主观的。从这两类方法中获得的度量代表了视频的不同内在特征(Eude, T. and Mayache, a .， Proc. ICSP' 1998,1998)。然而，我们对各种视频进行了客观的质量评估，这些视频从无运动到高运动，由三种编码器(MPEG-1, MPEG-2和Windows Media格式)以两种不同的帧速率，两种帧大小和三种低比特率组合编码，以确定编码和解码过程中引入的视频内容的变化。研究了编码、码率、帧率和帧大小对视频质量的影响。我们定义并使用了一个质量度量，即累积亮度误差。它是每帧颜色分量的累积均方误差，对整个视频序列进行平均。它支持基于内容差异和编码参数值的视频格式比较。
{"title":"Objective quality analysis of MPEG-1, MPEG-2 & Windows Media video","authors":"P. Aeluri, V. Bojan, S. Richie, A. Weeks","doi":"10.1109/IAI.2004.1300978","DOIUrl":"https://doi.org/10.1109/IAI.2004.1300978","url":null,"abstract":"Video quality estimation methods in the literature can be broadly classified into two classes: objective and subjective. The metrics obtained from both classes of methods represent different intrinsic features of a video (Eude, T. and Mayache, A., Proc. ICSP'98, 1998). However, we have performed an objective quality evaluation on various videos, which range from no motion to high motion, encoded by three encoders (MPEG-1, MPEG-2 and Windows Media formats) at two different frame rates, two frame sizes and three low bit rate combinations for the purpose of determining the change in video content introduced by the encoding and decoding process. The effect of the encoding, bit rates, frame rates and frame sizes on the quality of videos has been studied. We have defined and used a quality metric, cumulative brightness error. It is the cumulative mean square error of each frame's color components, averaged for the entire video sequence. It enables comparison of video formats based on content differences and encoding parameter values.","PeriodicalId":326040,"journal":{"name":"6th IEEE Southwest Symposium on Image Analysis and Interpretation, 2004.","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2004-03-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132082690","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：