Pub Date : 2001-09-26DOI: 10.1109/ICIAP.2001.957032
Giuseppe Boccignone, M. Ferraro, T. Caelli
An approach to active vision based on information theory and statistical mechanics is presented. Density of entropy production measured along a spatio-chromatic diffusion of a colour image is used to build a conspicuity map of the image. The map is successively given as input to a dynamic neural network in order to drive a focus-of-attention scanpath.
{"title":"An information-theoretic approach to active vision","authors":"Giuseppe Boccignone, M. Ferraro, T. Caelli","doi":"10.1109/ICIAP.2001.957032","DOIUrl":"https://doi.org/10.1109/ICIAP.2001.957032","url":null,"abstract":"An approach to active vision based on information theory and statistical mechanics is presented. Density of entropy production measured along a spatio-chromatic diffusion of a colour image is used to build a conspicuity map of the image. The map is successively given as input to a dynamic neural network in order to drive a focus-of-attention scanpath.","PeriodicalId":365627,"journal":{"name":"Proceedings 11th International Conference on Image Analysis and Processing","volume":"12 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116786220","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 : 2001-09-26DOI: 10.1109/ICIAP.2001.957053
Bin Luo, E. Hancock
Graph-matching is a task of pivotal importance in high-level vision since it provides a means by which abstract pictorial descriptions can be matched to one another. This paper describes an efficient algorithm for inexact graph-matching. The method is purely structural, that is to say it uses only the edge or connectivity structure of the graph and does not draw on node or edge attributes. We make two contributions. Commencing from a probability distribution for matching errors, we show how the problem of graph-matching can be posed as maximum likelihood estimation using the apparatus of the EM algorithm. Our second contribution is to cast the recovery of correspondence matches between the graph nodes in a matrix framework. This allows us to efficiently recover correspondence matches using singular value decomposition. We experiment with the method on both real-world and synthetic data. Here we demonstrate that the method offers comparable performance to more computationally demanding methods.
{"title":"A robust eigendecomposition framework for inexact graph-matching","authors":"Bin Luo, E. Hancock","doi":"10.1109/ICIAP.2001.957053","DOIUrl":"https://doi.org/10.1109/ICIAP.2001.957053","url":null,"abstract":"Graph-matching is a task of pivotal importance in high-level vision since it provides a means by which abstract pictorial descriptions can be matched to one another. This paper describes an efficient algorithm for inexact graph-matching. The method is purely structural, that is to say it uses only the edge or connectivity structure of the graph and does not draw on node or edge attributes. We make two contributions. Commencing from a probability distribution for matching errors, we show how the problem of graph-matching can be posed as maximum likelihood estimation using the apparatus of the EM algorithm. Our second contribution is to cast the recovery of correspondence matches between the graph nodes in a matrix framework. This allows us to efficiently recover correspondence matches using singular value decomposition. We experiment with the method on both real-world and synthetic data. Here we demonstrate that the method offers comparable performance to more computationally demanding methods.","PeriodicalId":365627,"journal":{"name":"Proceedings 11th International Conference on Image Analysis and Processing","volume":"160 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115569708","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 : 2001-09-26DOI: 10.1109/ICIAP.2001.957069
A. Chella, Salvatore Vitabile, R. Sorbello
We present an architecture for mobile robot navigation based on Bayesian networks. The architecture allows a robot to plan the correct path inside an environment with dynamic obstacles. Interactions between the robot and the environment are based on a powerful vision agent. The results of simulations, showing the effectiveness of the approach, are described.
{"title":"A vision agent for mobile robot navigation in time-variable environments","authors":"A. Chella, Salvatore Vitabile, R. Sorbello","doi":"10.1109/ICIAP.2001.957069","DOIUrl":"https://doi.org/10.1109/ICIAP.2001.957069","url":null,"abstract":"We present an architecture for mobile robot navigation based on Bayesian networks. The architecture allows a robot to plan the correct path inside an environment with dynamic obstacles. Interactions between the robot and the environment are based on a powerful vision agent. The results of simulations, showing the effectiveness of the approach, are described.","PeriodicalId":365627,"journal":{"name":"Proceedings 11th International Conference on Image Analysis and Processing","volume":"25 4 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"116573837","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 : 2001-09-26DOI: 10.1109/ICIAP.2001.957052
Jussi Tohka
Deformable models are by their formulation able to solve the surface extraction problem from noisy volumetric image data encountered commonly in medical image analysis. However, this ability is shadowed by the fact that the minimization problem formulated is difficult to solve globally. Constrained global solutions are needed, if the amount of noise is substantial. This paper presents a new optimization strategy for deformable surface meshes based on real coded genetic algorithms. Real coded genetic algorithms are favored over binary coded ones because they can more efficiently be adapted to the particular problem domain. Experiments with synthetic images are performed. These demonstrate that the applied deformable model is able extract a surface from noisy volumetric image. Also the superiority of the proposed approach compared to a greedy minimization with multiple initializations is demonstrated.
{"title":"Global optimization of deformable surface meshes based on genetic algorithms","authors":"Jussi Tohka","doi":"10.1109/ICIAP.2001.957052","DOIUrl":"https://doi.org/10.1109/ICIAP.2001.957052","url":null,"abstract":"Deformable models are by their formulation able to solve the surface extraction problem from noisy volumetric image data encountered commonly in medical image analysis. However, this ability is shadowed by the fact that the minimization problem formulated is difficult to solve globally. Constrained global solutions are needed, if the amount of noise is substantial. This paper presents a new optimization strategy for deformable surface meshes based on real coded genetic algorithms. Real coded genetic algorithms are favored over binary coded ones because they can more efficiently be adapted to the particular problem domain. Experiments with synthetic images are performed. These demonstrate that the applied deformable model is able extract a surface from noisy volumetric image. Also the superiority of the proposed approach compared to a greedy minimization with multiple initializations is demonstrated.","PeriodicalId":365627,"journal":{"name":"Proceedings 11th International Conference on Image Analysis and Processing","volume":"65 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"128685766","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 : 2001-09-26DOI: 10.1109/ICIAP.2001.957059
V. Roberto, V. D. Mea
A network of distributed image-based medical services (Imagencies) is proposed. Each one is a community of specialised, co-operating programs (Imagents), providing acquisition, processing, archiving and management facilities. Both synchronous and asynchronous communication are proposed, based on the standard protocols available on the Internet. We describe a prototype implementation in the field of telepathology. At present, the Imagency includes a secretary and an image processing specialist; asynchronous communication is realised by the standard MIME protocol for multimedia electronic mail. Practical results and future perspectives are presented and discussed.
{"title":"IMAGENCIES: network image services for telemedicine applications","authors":"V. Roberto, V. D. Mea","doi":"10.1109/ICIAP.2001.957059","DOIUrl":"https://doi.org/10.1109/ICIAP.2001.957059","url":null,"abstract":"A network of distributed image-based medical services (Imagencies) is proposed. Each one is a community of specialised, co-operating programs (Imagents), providing acquisition, processing, archiving and management facilities. Both synchronous and asynchronous communication are proposed, based on the standard protocols available on the Internet. We describe a prototype implementation in the field of telepathology. At present, the Imagency includes a secretary and an image processing specialist; asynchronous communication is realised by the standard MIME protocol for multimedia electronic mail. Practical results and future perspectives are presented and discussed.","PeriodicalId":365627,"journal":{"name":"Proceedings 11th International Conference on Image Analysis and Processing","volume":"361 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"130335880","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 : 2001-09-26DOI: 10.1109/ICIAP.2001.957050
H. Pedrini
A method for approximating range images by integrating triangular meshes and curvature information is presented. First, an adaptive filtering technique is applied to the original range image based on estimations of the surface curvature. This produces a collection of 3D points, which are triangulated in order to produce an initial mesh. The mesh is then refined through an efficient Delaunay triangulation algorithm. A new local error measure is used to select points to be inserted into the triangulation. Points tend to scatter in planar areas and to concentrate in high variation areas. The method allows representations to be retrieved at variables levels of accuracy, providing a natural way of multiresolution modeling. Some experimental results are presented to show that the proposed technique is effective to represent range images.
{"title":"Modeling dense range images through fast polygonal approximations","authors":"H. Pedrini","doi":"10.1109/ICIAP.2001.957050","DOIUrl":"https://doi.org/10.1109/ICIAP.2001.957050","url":null,"abstract":"A method for approximating range images by integrating triangular meshes and curvature information is presented. First, an adaptive filtering technique is applied to the original range image based on estimations of the surface curvature. This produces a collection of 3D points, which are triangulated in order to produce an initial mesh. The mesh is then refined through an efficient Delaunay triangulation algorithm. A new local error measure is used to select points to be inserted into the triangulation. Points tend to scatter in planar areas and to concentrate in high variation areas. The method allows representations to be retrieved at variables levels of accuracy, providing a natural way of multiresolution modeling. Some experimental results are presented to show that the proposed technique is effective to represent range images.","PeriodicalId":365627,"journal":{"name":"Proceedings 11th International Conference on Image Analysis and Processing","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129173192","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 : 2001-09-26DOI: 10.1109/ICIAP.2001.956991
R. Palenichka, M. Zaremba, C. Valenti
This paper describes a fast algorithm to compute local axial moments used for the detection of objects of interest in images. The basic idea is grounded on the elimination of redundant operations while computing axial moments for two neighboring angles of orientation. The main result is that the complexity of recursive computation of axial moments becomes independent of the total number of computed moments in a given point, i.e. it is of the order O(N) where N is the data size. This result is of great importance in computer vision since many feature extraction methods are based on the computation of axial moments. The experimental results confirm the time complexity and accuracy predicted by the theoretical analysis.
{"title":"A fast recursive algorithm for the computation of axial moments","authors":"R. Palenichka, M. Zaremba, C. Valenti","doi":"10.1109/ICIAP.2001.956991","DOIUrl":"https://doi.org/10.1109/ICIAP.2001.956991","url":null,"abstract":"This paper describes a fast algorithm to compute local axial moments used for the detection of objects of interest in images. The basic idea is grounded on the elimination of redundant operations while computing axial moments for two neighboring angles of orientation. The main result is that the complexity of recursive computation of axial moments becomes independent of the total number of computed moments in a given point, i.e. it is of the order O(N) where N is the data size. This result is of great importance in computer vision since many feature extraction methods are based on the computation of axial moments. The experimental results confirm the time complexity and accuracy predicted by the theoretical analysis.","PeriodicalId":365627,"journal":{"name":"Proceedings 11th International Conference on Image Analysis and Processing","volume":"16 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"123883230","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 : 2001-09-26DOI: 10.1109/ICIAP.2001.956998
G. Gallo, Giorgio Grasso, Salvatore Nicotra, A. Pulvirenti
A novel approach to the automatic classification of remotely sensed images is proposed. This approach is based on a three-phase procedure: first pixels which belong to the areas of interest with large likelihood are selected as seeds; second the seeds are refined into connected shapes using two well-known image processing techniques; third the results of the shape refinement algorithms are merged together. The initial seed extraction is performed using a simple thresholding strategy applied to NDVI/sub 4-3/ index. Subsequently shape refinement through seeded region growing and watershed decomposition is applied; finally a merging procedure is applied to build likelihood maps. Experimental results are presented to analyze the correctness and robustness of the method in recognizing vegetation areas around Mount Etna.
{"title":"Remote sensed images segmentation through shape refinement","authors":"G. Gallo, Giorgio Grasso, Salvatore Nicotra, A. Pulvirenti","doi":"10.1109/ICIAP.2001.956998","DOIUrl":"https://doi.org/10.1109/ICIAP.2001.956998","url":null,"abstract":"A novel approach to the automatic classification of remotely sensed images is proposed. This approach is based on a three-phase procedure: first pixels which belong to the areas of interest with large likelihood are selected as seeds; second the seeds are refined into connected shapes using two well-known image processing techniques; third the results of the shape refinement algorithms are merged together. The initial seed extraction is performed using a simple thresholding strategy applied to NDVI/sub 4-3/ index. Subsequently shape refinement through seeded region growing and watershed decomposition is applied; finally a merging procedure is applied to build likelihood maps. Experimental results are presented to analyze the correctness and robustness of the method in recognizing vegetation areas around Mount Etna.","PeriodicalId":365627,"journal":{"name":"Proceedings 11th International Conference on Image Analysis and Processing","volume":"560 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"134194505","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 : 2001-09-26DOI: 10.1109/ICIAP.2001.956981
B. Moghaddam, C. Nastar, A. Pentland
We propose a novel representation for characterizing image differences using a deformable technique for obtaining pixel-wise correspondences. This representation, which is based on a deformable 3D mesh in XYI-space, is then experimentally compared with two related correspondence methods: optical flow and intensity differences. Furthermore, we make use of a probabilistic similarity measure for direct image matching based on a Bayesian analysis of image variations. We model two classes of variation in facial appearance: intra-personal and extra-personal. The probability density function for each class is estimated from training data and used to compute a similarity measure based on the a posteriori probabilities. The performance advantage of our deformable probabilistic matching technique is demonstrated using 1700 faces from the USA Army's "FERET" face database.
{"title":"Bayesian face recognition with deformable image models","authors":"B. Moghaddam, C. Nastar, A. Pentland","doi":"10.1109/ICIAP.2001.956981","DOIUrl":"https://doi.org/10.1109/ICIAP.2001.956981","url":null,"abstract":"We propose a novel representation for characterizing image differences using a deformable technique for obtaining pixel-wise correspondences. This representation, which is based on a deformable 3D mesh in XYI-space, is then experimentally compared with two related correspondence methods: optical flow and intensity differences. Furthermore, we make use of a probabilistic similarity measure for direct image matching based on a Bayesian analysis of image variations. We model two classes of variation in facial appearance: intra-personal and extra-personal. The probability density function for each class is estimated from training data and used to compute a similarity measure based on the a posteriori probabilities. The performance advantage of our deformable probabilistic matching technique is demonstrated using 1700 faces from the USA Army's \"FERET\" face database.","PeriodicalId":365627,"journal":{"name":"Proceedings 11th International Conference on Image Analysis and Processing","volume":"6 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115323848","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 : 2001-09-26DOI: 10.1109/ICIAP.2001.957016
P. Parada, Javier Ruiz-del-Solar
Markov random field (MRF) is a well known model for the generation of textures in the field of computer graphics. However, the estimation of its parameter is quite difficult in many cases. A new algorithm for the synthesis of textures is proposed, based on image pyramids and self-organizing maps. This procedure avoids the explicit computation of the MRF parameters. Preliminary results support the appropriateness of this new approach.
{"title":"Texture synthesis using image pyramids and self-organizing maps","authors":"P. Parada, Javier Ruiz-del-Solar","doi":"10.1109/ICIAP.2001.957016","DOIUrl":"https://doi.org/10.1109/ICIAP.2001.957016","url":null,"abstract":"Markov random field (MRF) is a well known model for the generation of textures in the field of computer graphics. However, the estimation of its parameter is quite difficult in many cases. A new algorithm for the synthesis of textures is proposed, based on image pyramids and self-organizing maps. This procedure avoids the explicit computation of the MRF parameters. Preliminary results support the appropriateness of this new approach.","PeriodicalId":365627,"journal":{"name":"Proceedings 11th International Conference on Image Analysis and Processing","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2001-09-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124724201","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}