Pub Date : 2016-10-01DOI: 10.1109/ICCAIS.2016.7822438
Feng Yang, Wanying Zhang, Yan Liang, Xiaoxu Wang, Linfeng Xu
Cardinality balanced multi-target multi-Bernoulli (CBMeMBer) filter has been proved as a promising method in the context of multi-target tracking with an unknown number of targets, clutter and false alarms. For tracking maneuvering targets, the CBMeMBer filter has been extended by using jump Markov models (JMM). However, the standard particle implementation of the multiple model CBMeMBer (MM-CBMeMBer) filter requires a large number of particles in order to obtain a satisfactory performance. Based on the capability of box-particle filter to process measurements which are affected by bounded errors of unknown distributions and biases, a box-particle implementation of the MM-CBMeMBer filter is proposed. Simulation result shows that the proposed MM-Box-CBMeMBer filter can obtain similar accuracy results with a MM-Particle-CBMeMBer filter but considerably reduce the computational costs. Meanwhile, in the presence of strongly biased measurements, it is shown that the MM-Box-CBMeMBer filter is superior to the MM-Particle-CBMeMBer filter.
{"title":"Multiple model box-particle cardinality balanced multi-target multi-Bernoulli filter for multiple maneuvering targets tracking","authors":"Feng Yang, Wanying Zhang, Yan Liang, Xiaoxu Wang, Linfeng Xu","doi":"10.1109/ICCAIS.2016.7822438","DOIUrl":"https://doi.org/10.1109/ICCAIS.2016.7822438","url":null,"abstract":"Cardinality balanced multi-target multi-Bernoulli (CBMeMBer) filter has been proved as a promising method in the context of multi-target tracking with an unknown number of targets, clutter and false alarms. For tracking maneuvering targets, the CBMeMBer filter has been extended by using jump Markov models (JMM). However, the standard particle implementation of the multiple model CBMeMBer (MM-CBMeMBer) filter requires a large number of particles in order to obtain a satisfactory performance. Based on the capability of box-particle filter to process measurements which are affected by bounded errors of unknown distributions and biases, a box-particle implementation of the MM-CBMeMBer filter is proposed. Simulation result shows that the proposed MM-Box-CBMeMBer filter can obtain similar accuracy results with a MM-Particle-CBMeMBer filter but considerably reduce the computational costs. Meanwhile, in the presence of strongly biased measurements, it is shown that the MM-Box-CBMeMBer filter is superior to the MM-Particle-CBMeMBer filter.","PeriodicalId":407031,"journal":{"name":"2016 International Conference on Control, Automation and Information Sciences (ICCAIS)","volume":"28 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114286829","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 : 2016-10-01DOI: 10.1109/ICCAIS.2016.7822443
Phairod Tuntiwachiratrakun, P. Vateekul
Nowadays, classification tasks are very challenging because data is usually large and imbalanced. They can cause low prediction accuracy and high computation costs. Active Learning is a technique that employs only a small set of data to construct an initial classification model. Then, it iteratively improves the model by incrementally learning from the misclassified examples. In this paper, we aim to improve prediction accuracy by applying Active Learning. To solve the imbalance issue, the active model was iteratively updated based on the G-mean, and the under sampling sampling was also applied. The proposed algorithm was suitable for large scale data since it did not need to use the whole data set to construct a model. The experiment was conducted on two standard corpuses, one of which contained more than 100,000 examples. The result showed that a prediction performance of standard technique (Neural Network) can be improved by applying the Active Learning strategy for 5%–13%. Furthermore, this technique also outperformed other classical classification algorithms including K-nearest neighbors (kNN), Support Vector Machine (SVM), Decision Tree (DT), Naïve Bayes (NB) and Artificial Neural Network (ANN).
{"title":"Applying active learning strategy to classify large scale data with imbalanced classes","authors":"Phairod Tuntiwachiratrakun, P. Vateekul","doi":"10.1109/ICCAIS.2016.7822443","DOIUrl":"https://doi.org/10.1109/ICCAIS.2016.7822443","url":null,"abstract":"Nowadays, classification tasks are very challenging because data is usually large and imbalanced. They can cause low prediction accuracy and high computation costs. Active Learning is a technique that employs only a small set of data to construct an initial classification model. Then, it iteratively improves the model by incrementally learning from the misclassified examples. In this paper, we aim to improve prediction accuracy by applying Active Learning. To solve the imbalance issue, the active model was iteratively updated based on the G-mean, and the under sampling sampling was also applied. The proposed algorithm was suitable for large scale data since it did not need to use the whole data set to construct a model. The experiment was conducted on two standard corpuses, one of which contained more than 100,000 examples. The result showed that a prediction performance of standard technique (Neural Network) can be improved by applying the Active Learning strategy for 5%–13%. Furthermore, this technique also outperformed other classical classification algorithms including K-nearest neighbors (kNN), Support Vector Machine (SVM), Decision Tree (DT), Naïve Bayes (NB) and Artificial Neural Network (ANN).","PeriodicalId":407031,"journal":{"name":"2016 International Conference on Control, Automation and Information Sciences (ICCAIS)","volume":"39 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"124138663","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 : 2016-10-01DOI: 10.1109/ICCAIS.2016.7822446
H. T. Nguyen, P. Tran
Flyer is a means utilized popularly and effectively in recording landscape from above, collecting data of environment and weather, contributing to field works, especially conveying samples and medications from or to patients living in areas isolated by flood. It is necessary to track flyer's locations and characteristics for its activities. Visualization-based system for tracking flyer enables user to monitor flyer by analyzing visually multivariable flight data. Visualization-based system for tracking flyer using 3D cube represents flight data including ground position and elevation, using space-time cube (STC) represents ground position and time, using 4D cube represents time, ground position, and elevation. Meanwhile, multipurpose flyer needs to be tracked not only time, ground position, elevation, but also characteristics. The paper proposes multidimensional cube (mD cube) for visualization-based tracking system to represent multivariable flight data including time, location, and characteristics. Multidimensional cube results from the combination of a 4D cube with a multivariate cube representing characteristics changing over time. The mD cube represents visually multivariable flight data in visualization-based tracking system to enable user to monitor flyer. With mathematical reasoning, user can understand the significance of multivariable flight data by responding several analytical tasks.
{"title":"Multidimensional cube for representing flight data in visualization-based system for tracking flyer","authors":"H. T. Nguyen, P. Tran","doi":"10.1109/ICCAIS.2016.7822446","DOIUrl":"https://doi.org/10.1109/ICCAIS.2016.7822446","url":null,"abstract":"Flyer is a means utilized popularly and effectively in recording landscape from above, collecting data of environment and weather, contributing to field works, especially conveying samples and medications from or to patients living in areas isolated by flood. It is necessary to track flyer's locations and characteristics for its activities. Visualization-based system for tracking flyer enables user to monitor flyer by analyzing visually multivariable flight data. Visualization-based system for tracking flyer using 3D cube represents flight data including ground position and elevation, using space-time cube (STC) represents ground position and time, using 4D cube represents time, ground position, and elevation. Meanwhile, multipurpose flyer needs to be tracked not only time, ground position, elevation, but also characteristics. The paper proposes multidimensional cube (mD cube) for visualization-based tracking system to represent multivariable flight data including time, location, and characteristics. Multidimensional cube results from the combination of a 4D cube with a multivariate cube representing characteristics changing over time. The mD cube represents visually multivariable flight data in visualization-based tracking system to enable user to monitor flyer. With mathematical reasoning, user can understand the significance of multivariable flight data by responding several analytical tasks.","PeriodicalId":407031,"journal":{"name":"2016 International Conference on Control, Automation and Information Sciences (ICCAIS)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129964891","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 : 2016-10-01DOI: 10.1109/ICCAIS.2016.7822462
Fei Gao, Mingli Lu, Benlian Xu, Qian Zhang
This paper is aimed at the difficult problem of multi region segmentation of weld pool image, analyzed The difficulty of edge extraction in the inner region of the weld pool. According to the characteristics between pixel neighborhood space and neighbor pixel correlation, based on local standard deviation, presented a noise suppression, edge enhancement of the weld pool image multi region division and multi region edge detection algorithm, Through the test of the weld pool image, It shows that the algorithm can accurately divide the internal details of the weld pool. Finally, the Sobel operator, Roberts operator, Prewitt operator and the edge detection results of the weld pool image are analyzed and compared by experiments, The results show that the algorithm in this paper is much better than other algorithms, At last, the accuracy of the algorithm is tested by the difference shadow detection, a continuous multi region edge was obtained by the expansion of corrosion.
{"title":"Multi region segmentation algorithm based on edge preserving for molten pool image","authors":"Fei Gao, Mingli Lu, Benlian Xu, Qian Zhang","doi":"10.1109/ICCAIS.2016.7822462","DOIUrl":"https://doi.org/10.1109/ICCAIS.2016.7822462","url":null,"abstract":"This paper is aimed at the difficult problem of multi region segmentation of weld pool image, analyzed The difficulty of edge extraction in the inner region of the weld pool. According to the characteristics between pixel neighborhood space and neighbor pixel correlation, based on local standard deviation, presented a noise suppression, edge enhancement of the weld pool image multi region division and multi region edge detection algorithm, Through the test of the weld pool image, It shows that the algorithm can accurately divide the internal details of the weld pool. Finally, the Sobel operator, Roberts operator, Prewitt operator and the edge detection results of the weld pool image are analyzed and compared by experiments, The results show that the algorithm in this paper is much better than other algorithms, At last, the accuracy of the algorithm is tested by the difference shadow detection, a continuous multi region edge was obtained by the expansion of corrosion.","PeriodicalId":407031,"journal":{"name":"2016 International Conference on Control, Automation and Information Sciences (ICCAIS)","volume":"10 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2016-10-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"115251922","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：