Two class classification problems in real world are often characterized by imbalanced classes. This is a serious issue since a classifier trained on such a data distribution typically exhibits a prediction accuracy highly skewed towards the majority class. To improve the quality of the classifier, many approaches have been proposed till now for building artificially balanced training sets. Such methods are mainly based on undersampling the majority class and/or oversampling the minority class. However, both approaches can produce overfitting or underfitting problems for the trained classifier. In this paper we present a method for building a multiple classifier system in which each constituting classifier is trained on a subset of the majority class and on the whole minority class. The approach has been tested on the detection of microcalcifications on digital mammograms. The results obtained confirm the effectiveness of the method.
{"title":"Facing Imbalanced Classes through Aggregation of Classifiers","authors":"M. Molinara, M. Ricamato, F. Tortorella","doi":"10.1109/ICIAP.2007.65","DOIUrl":"https://doi.org/10.1109/ICIAP.2007.65","url":null,"abstract":"Two class classification problems in real world are often characterized by imbalanced classes. This is a serious issue since a classifier trained on such a data distribution typically exhibits a prediction accuracy highly skewed towards the majority class. To improve the quality of the classifier, many approaches have been proposed till now for building artificially balanced training sets. Such methods are mainly based on undersampling the majority class and/or oversampling the minority class. However, both approaches can produce overfitting or underfitting problems for the trained classifier. In this paper we present a method for building a multiple classifier system in which each constituting classifier is trained on a subset of the majority class and on the whole minority class. The approach has been tested on the detection of microcalcifications on digital mammograms. The results obtained confirm the effectiveness of the method.","PeriodicalId":118466,"journal":{"name":"14th International Conference on Image Analysis and Processing (ICIAP 2007)","volume":"35 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132399380","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}
Transformation of image patches is a common requirement for 2D transition animations such as shape interpolation and image morphing. It is usually done by applying affine transformations to triangular patches. However, the affine transformation does not model the perspective transformation frequently found in images. Hence, such techniques can only produce approximate results and usually use an excessively large number of triangles to compensate for this shortcoming. This paper proposes the application of projective transformations on quadrilateral image patches as a solution to this problem. We address the issues of appropriate decomposition and interpolation of projective transformation matrices to produce a natural looking transition animation for a single quadrilateral as well as for shapes made up of multiple quadrilaterals.
{"title":"Projective Transformations for Image Transition Animations","authors":"TzuYen Wong, P. Kovesi, A. Datta","doi":"10.1109/ICIAP.2007.103","DOIUrl":"https://doi.org/10.1109/ICIAP.2007.103","url":null,"abstract":"Transformation of image patches is a common requirement for 2D transition animations such as shape interpolation and image morphing. It is usually done by applying affine transformations to triangular patches. However, the affine transformation does not model the perspective transformation frequently found in images. Hence, such techniques can only produce approximate results and usually use an excessively large number of triangles to compensate for this shortcoming. This paper proposes the application of projective transformations on quadrilateral image patches as a solution to this problem. We address the issues of appropriate decomposition and interpolation of projective transformation matrices to produce a natural looking transition animation for a single quadrilateral as well as for shapes made up of multiple quadrilaterals.","PeriodicalId":118466,"journal":{"name":"14th International Conference on Image Analysis and Processing (ICIAP 2007)","volume":"30 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132013632","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}
An essential component of new advanced intelligent video surveillance systems is the possibility to perform non-cooperative people detection and identification. Nowadays Face Detection and Recognition are used in access control systems where the human being is willing to help the system, but the task is much more complex in other unconstrained situations. The paper refers some results of a preliminary study to investigate the most promising approach for face detection in noncooperative conditions with the main objective to reduce as much as possible the number of false alarms working on video-rate processing speed. The proposed solution has been developed around the AdaBoost approach, using the open-CV library, with an integration of motion and colour segmentation. The primary scope of the paper is to refer some experimental results to show the potential improvements in terms of reduction of false positives and a significant decrease of the execution time.
{"title":"An Improvement of AdaBoost for Face-Detection with Motion and Color Information","authors":"V. Randazzo, Lisa Usai","doi":"10.1109/ICIAP.2007.22","DOIUrl":"https://doi.org/10.1109/ICIAP.2007.22","url":null,"abstract":"An essential component of new advanced intelligent video surveillance systems is the possibility to perform non-cooperative people detection and identification. Nowadays Face Detection and Recognition are used in access control systems where the human being is willing to help the system, but the task is much more complex in other unconstrained situations. The paper refers some results of a preliminary study to investigate the most promising approach for face detection in noncooperative conditions with the main objective to reduce as much as possible the number of false alarms working on video-rate processing speed. The proposed solution has been developed around the AdaBoost approach, using the open-CV library, with an integration of motion and colour segmentation. The primary scope of the paper is to refer some experimental results to show the potential improvements in terms of reduction of false positives and a significant decrease of the execution time.","PeriodicalId":118466,"journal":{"name":"14th International Conference on Image Analysis and Processing (ICIAP 2007)","volume":"45 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132172372","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}
We address the problem of recognizing the so-called image spam, which consists in embedding the spam message into attached images to defeat techniques based on the analysis of e-mails' body text, and in using content obscuring techniques to defeat OCR tools. We propose an approach to recognize image spam based on detecting the presence of content obscuring techniques, and describe a possible implementation based on two low-level image features aimed at detecting obscuring techniques whose consequence is to compromise the OCR effectiveness resulting in character breaking or merging, or in the presence of noise interfering with characters in the binarized image. A preliminary experimental investigation of this approach is reported on a personal data set of spam images.
{"title":"Image Spam Filtering Using Visual Information","authors":"B. Biggio, G. Fumera, I. Pillai, F. Roli","doi":"10.1109/ICIAP.2007.79","DOIUrl":"https://doi.org/10.1109/ICIAP.2007.79","url":null,"abstract":"We address the problem of recognizing the so-called image spam, which consists in embedding the spam message into attached images to defeat techniques based on the analysis of e-mails' body text, and in using content obscuring techniques to defeat OCR tools. We propose an approach to recognize image spam based on detecting the presence of content obscuring techniques, and describe a possible implementation based on two low-level image features aimed at detecting obscuring techniques whose consequence is to compromise the OCR effectiveness resulting in character breaking or merging, or in the presence of noise interfering with characters in the binarized image. A preliminary experimental investigation of this approach is reported on a personal data set of spam images.","PeriodicalId":118466,"journal":{"name":"14th International Conference on Image Analysis and Processing (ICIAP 2007)","volume":"55 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125519013","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}
In this paper, a novel learning algorithm for Hidden Markov Models (HMMs) has been devised. The key issue is the achievement of a sparse model, i.e., a model in which all irrelevant parameters are set exactly to zero. Alternatively to standard maximum likelihood estimation (Baum Welch training), in the proposed approach the parameters estimation problem is cast into a Bayesian framework, with the introduction of a negative Dirichlet prior, which strongly encourages sparseness of the model. A modified Expectation Maximization algorithm has been devised, able to determine a MAP (maximum a posteriori probability) estimate of HMM parameters in this Bayesian formulation. Theoretical considerations and experimental comparative evaluations on a 2D shape classification task contribute to validate the proposed technique.
{"title":"Sparseness Achievement in Hidden Markov Models","authors":"M. Bicego, M. Cristani, Vittorio Murino","doi":"10.1109/ICIAP.2007.118","DOIUrl":"https://doi.org/10.1109/ICIAP.2007.118","url":null,"abstract":"In this paper, a novel learning algorithm for Hidden Markov Models (HMMs) has been devised. The key issue is the achievement of a sparse model, i.e., a model in which all irrelevant parameters are set exactly to zero. Alternatively to standard maximum likelihood estimation (Baum Welch training), in the proposed approach the parameters estimation problem is cast into a Bayesian framework, with the introduction of a negative Dirichlet prior, which strongly encourages sparseness of the model. A modified Expectation Maximization algorithm has been devised, able to determine a MAP (maximum a posteriori probability) estimate of HMM parameters in this Bayesian formulation. Theoretical considerations and experimental comparative evaluations on a 2D shape classification task contribute to validate the proposed technique.","PeriodicalId":118466,"journal":{"name":"14th International Conference on Image Analysis and Processing (ICIAP 2007)","volume":"14 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126242685","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}
P. Gemeiner, W. Ponweiser, P. Einramhof, M. Vincze
A typical task in mobile robotics or augmented reality applications is self-localization in an unknown environment. In the robotics community the localization and mapping of the unknown environment is called SLAM (simultaneous localization and mapping). Important constraints in SLAM using visual input are real-time processing and robustness against motion blur or jitter. The contribution of this paper is in enhancing the performance of a well known SLAM method using a high-speed CMOS camera. Benefits of this camera are that it allows fast image processing, little motion blur and low localization uncertainty. SLAM performance with a high-speed camera is demonstrated on a robotic arm. The result of the experiments is that for SLAM applications in robotics, where the motion is not smooth, the high-speed CMOS camera is a more suitable sensor than a standard CCD camera.
{"title":"Real-Time SLAM with a High-Speed CMOS Camera","authors":"P. Gemeiner, W. Ponweiser, P. Einramhof, M. Vincze","doi":"10.1109/ICIAP.2007.108","DOIUrl":"https://doi.org/10.1109/ICIAP.2007.108","url":null,"abstract":"A typical task in mobile robotics or augmented reality applications is self-localization in an unknown environment. In the robotics community the localization and mapping of the unknown environment is called SLAM (simultaneous localization and mapping). Important constraints in SLAM using visual input are real-time processing and robustness against motion blur or jitter. The contribution of this paper is in enhancing the performance of a well known SLAM method using a high-speed CMOS camera. Benefits of this camera are that it allows fast image processing, little motion blur and low localization uncertainty. SLAM performance with a high-speed camera is demonstrated on a robotic arm. The result of the experiments is that for SLAM applications in robotics, where the motion is not smooth, the high-speed CMOS camera is a more suitable sensor than a standard CCD camera.","PeriodicalId":118466,"journal":{"name":"14th International Conference on Image Analysis and Processing (ICIAP 2007)","volume":"27 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124919575","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}
S. Mattoccia, Federico Tombari, L. D. Stefano, Marco Pignoloni
This paper proposes a novel technique for performing fast block matching for motion estimation which is optimal, meaning it yields the same results as a full-search investigation. The proposed technique derives from an approach previously proposed (Tombari et al., 2006) for template matching and it is based on the deployment of a succession of lower bounding functions of the matching metric. Hence, an algorithm is outlined which efficiently exploits these bounding functions in order to rapidly determine non-matching block candidates, thus reducing the overall computational burden. Experimental results show that, compared to the brute-force approach, the proposed technique allows for notable reductions in terms of number of operations and computation times.
本文提出了一种用于运动估计的快速块匹配的新技术，该技术是最优的，这意味着它产生与全搜索调查相同的结果。所提出的技术源自先前提出的模板匹配方法(Tombari et al.， 2006)，它基于匹配度量的一系列下限函数的部署。因此，本文提出了一种有效利用这些边界函数的算法，以快速确定不匹配的候选块，从而减少总体计算负担。实验结果表明，与暴力破解方法相比，该方法可以显著减少操作次数和计算时间。
{"title":"Efficient and optimal block matching for motion estimation","authors":"S. Mattoccia, Federico Tombari, L. D. Stefano, Marco Pignoloni","doi":"10.1109/ICIAP.2007.58","DOIUrl":"https://doi.org/10.1109/ICIAP.2007.58","url":null,"abstract":"This paper proposes a novel technique for performing fast block matching for motion estimation which is optimal, meaning it yields the same results as a full-search investigation. The proposed technique derives from an approach previously proposed (Tombari et al., 2006) for template matching and it is based on the deployment of a succession of lower bounding functions of the matching metric. Hence, an algorithm is outlined which efficiently exploits these bounding functions in order to rapidly determine non-matching block candidates, thus reducing the overall computational burden. Experimental results show that, compared to the brute-force approach, the proposed technique allows for notable reductions in terms of number of operations and computation times.","PeriodicalId":118466,"journal":{"name":"14th International Conference on Image Analysis and Processing (ICIAP 2007)","volume":"11 ","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131433710","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}
Andrew D. Bagdanov, A. Bimbo, F. Dini, W. Nunziati
In particle filter-based visual trackers, dynamic velocity components are typically incorporated into the state update equations. In these cases, there is a risk that the uncertainty in the model update stage can become amplified in unexpected and undesirable ways, leading to erroneous behavior of the tracker. To deal with this problem, we propose a continuously adaptive approach to estimating uncertainty in the particle filter, one that balances the uncertainty in its static and dynamic elements. We provide quantitative performance evaluation of the resulting particle filter tracker on a set of ten video sequences. Results are reported in terms of a metric that can be used to objectively evaluate the performance of visual trackers. This metric is used to compare our modified particle filter tracker and the continuously adaptive mean shift tracker. Results show that the performance of the particle filter is significantly improved through adaptive parameter estimation, particularly in cases of occlusions and nonlinear target motion.
{"title":"Adaptive uncertainty estimation for particle filter-based trackers","authors":"Andrew D. Bagdanov, A. Bimbo, F. Dini, W. Nunziati","doi":"10.1109/ICIAP.2007.19","DOIUrl":"https://doi.org/10.1109/ICIAP.2007.19","url":null,"abstract":"In particle filter-based visual trackers, dynamic velocity components are typically incorporated into the state update equations. In these cases, there is a risk that the uncertainty in the model update stage can become amplified in unexpected and undesirable ways, leading to erroneous behavior of the tracker. To deal with this problem, we propose a continuously adaptive approach to estimating uncertainty in the particle filter, one that balances the uncertainty in its static and dynamic elements. We provide quantitative performance evaluation of the resulting particle filter tracker on a set of ten video sequences. Results are reported in terms of a metric that can be used to objectively evaluate the performance of visual trackers. This metric is used to compare our modified particle filter tracker and the continuously adaptive mean shift tracker. Results show that the performance of the particle filter is significantly improved through adaptive parameter estimation, particularly in cases of occlusions and nonlinear target motion.","PeriodicalId":118466,"journal":{"name":"14th International Conference on Image Analysis and Processing (ICIAP 2007)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131511081","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}
E. Ardizzone, O. Gambino, M. Cascia, Liliana Lo Presti, R. Pirrone
In this paper, a new multi-modal non-rigid registration technique for medical images is presented. Firstly, the registration problem is outlined and some of the most common approaches reported, then, the proposed algorithm is presented. The proposed technique is based on mutual information maximization and computes a deformation field through a suitable globally smoothed affine piecewise transformation. The algorithm has been conceived with particular attention to computational load and accuracy of results. Experimental results involving intra-patient, inter-patients and atlas images on brain CT and MR (T1, T2 and PD modalities) are reported.
{"title":"Multi-modal non-rigid registration of medical images based on mutual information maximization","authors":"E. Ardizzone, O. Gambino, M. Cascia, Liliana Lo Presti, R. Pirrone","doi":"10.1109/ICIAP.2007.90","DOIUrl":"https://doi.org/10.1109/ICIAP.2007.90","url":null,"abstract":"In this paper, a new multi-modal non-rigid registration technique for medical images is presented. Firstly, the registration problem is outlined and some of the most common approaches reported, then, the proposed algorithm is presented. The proposed technique is based on mutual information maximization and computes a deformation field through a suitable globally smoothed affine piecewise transformation. The algorithm has been conceived with particular attention to computational load and accuracy of results. Experimental results involving intra-patient, inter-patients and atlas images on brain CT and MR (T1, T2 and PD modalities) are reported.","PeriodicalId":118466,"journal":{"name":"14th International Conference on Image Analysis and Processing (ICIAP 2007)","volume":"18 2 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"131567740","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}
A strong requirement to come up with secure and user- friendly ways to authenticate and identify people, to safeguard their rights and interests, has probably been the main guiding force behind biometrics research. Though a vast amount of research has been done to recognize humans based on still images, the problem is still far from solved for unconstrained scenarios. This has led to an increased interest in using video for the task of biometric recognition. Not only does video provide more information, but also is more suitable for recognizing humans in general surveillance scenarios. Other than the multitude of still frames, video makes it possible to characterize biometrics based on inherent dynamics like gait which is not possible with still images. In this paper, we describe several recent algorithms to illustrate the usefulness of videos to identify humans. A brief discussion on remaining challenges is also included.
{"title":"Video Biometrics","authors":"R. Chellappa, G. Aggarwal","doi":"10.1109/ICIAP.2007.132","DOIUrl":"https://doi.org/10.1109/ICIAP.2007.132","url":null,"abstract":"A strong requirement to come up with secure and user- friendly ways to authenticate and identify people, to safeguard their rights and interests, has probably been the main guiding force behind biometrics research. Though a vast amount of research has been done to recognize humans based on still images, the problem is still far from solved for unconstrained scenarios. This has led to an increased interest in using video for the task of biometric recognition. Not only does video provide more information, but also is more suitable for recognizing humans in general surveillance scenarios. Other than the multitude of still frames, video makes it possible to characterize biometrics based on inherent dynamics like gait which is not possible with still images. In this paper, we describe several recent algorithms to illustrate the usefulness of videos to identify humans. A brief discussion on remaining challenges is also included.","PeriodicalId":118466,"journal":{"name":"14th International Conference on Image Analysis and Processing (ICIAP 2007)","volume":"15 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2007-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132793346","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：