The nature of statistical learning theory
The nature of statistical learning theory
Machine Learning
MetaCost: a general method for making classifiers cost-sensitive
KDD '99 Proceedings of the fifth ACM SIGKDD international conference on Knowledge discovery and data mining
Learning When Negative Examples Abound
ECML '97 Proceedings of the 9th European Conference on Machine Learning
Application of Panoramic Annular Lens for Motion Analysis Tasks: Surveillance and Smoke Detection
ICPR '00 Proceedings of the International Conference on Pattern Recognition - Volume 4
Editorial: special issue on learning from imbalanced data sets
ACM SIGKDD Explorations Newsletter - Special issue on learning from imbalanced datasets
The Smoke Detection for Early Fire-Alarming System Base on Video Processing
IIH-MSP '06 Proceedings of the 2006 International Conference on Intelligent Information Hiding and Multimedia
Active learning for class imbalance problem
SIGIR '07 Proceedings of the 30th annual international ACM SIGIR conference on Research and development in information retrieval
SMOTE: synthetic minority over-sampling technique
Journal of Artificial Intelligence Research
The foundations of cost-sensitive learning
IJCAI'01 Proceedings of the 17th international joint conference on Artificial intelligence - Volume 2
LIBSVM: A library for support vector machines
ACM Transactions on Intelligent Systems and Technology (TIST)
An overview of statistical learning theory
IEEE Transactions on Neural Networks
IEEE Transactions on Neural Networks
Hi-index | 0.00 |
Techniques used in video smoke detection systems have been discussed noticeably in past few years. With the advantage of early fire alarm in large or specific spaces such as studio and tunnels, the video-based smoke detection systems would not have time delay as conventional detectors. In contrast, how to reduce false alarm and increase the generalization ability is the key issue for such state-of-the-art systems. In this paper, examples consisting of features extracted from a real time video are collected for the training of a discriminating model. A prototype of support vector machine (SVM) is therefore introduced for the discriminating model with the capability in small sample size training and the good generalization ability. In order to reduce the false alarm, the prototype is then extended to a class-imbalanced learning model to deal with rarity of the positive class. A number of assuming data are used for imbalanced test to cope with the real world of fire safety. The technique is optimistic to enhance accuracy and reduce false alarm in video-based smoke systems.