Machine Learning
A decision-theoretic generalization of on-line learning and an application to boosting
Journal of Computer and System Sciences - Special issue: 26th annual ACM symposium on the theory of computing & STOC'94, May 23–25, 1994, and second annual Europe an conference on computational learning theory (EuroCOLT'95), March 13–15, 1995
Decision Tree Induction Based on Efficient Tree Restructuring
Machine Learning
Combining labeled and unlabeled data with co-training
COLT' 98 Proceedings of the eleventh annual conference on Computational learning theory
Machine Learning
CVPR '05 Proceedings of the 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05) - Volume 2 - Volume 02
CVPR '06 Proceedings of the 2006 IEEE Computer Society Conference on Computer Vision and Pattern Recognition - Volume 1
On-line machine vision system for detect split defects in sheet-metal forming processes
ICPR '06 Proceedings of the 18th International Conference on Pattern Recognition - Volume 01
Fabric defect detection using modified local binary patterns
EURASIP Journal on Advances in Signal Processing
Online Manifold Regularization: A New Learning Setting and Empirical Study
ECML PKDD '08 Proceedings of the 2008 European Conference on Machine Learning and Knowledge Discovery in Databases - Part I
Semi-supervised On-Line Boosting for Robust Tracking
ECCV '08 Proceedings of the 10th European Conference on Computer Vision: Part I
On-line evolving image classifiers and their application to surface inspection
Image and Vision Computing
Image and Vision Computing
An unsupervised, online learning framework for moving object detection
CVPR'04 Proceedings of the 2004 IEEE computer society conference on Computer vision and pattern recognition
Hi-index | 0.00 |
Machine learning techniques are being used extensively for knowledge discovery and data mining in industrial inspection applications. Traditionally, all of the training samples are required to be presented for training a classifier at a batch mode. However, in most industrial inspection applications, only a small amount of training samples are available initially and larger amount of them become sequentially available during online prediction. In addition, data properties could be changing over time. We propose supervised and semi-supervised online learning with a Boosting Tree (BT) to adapt and evolve the classifier in an online fashion and thus accommodate new information that becomes available sequentially in industrial inspection applications. The supervised online BT can efficiently expand and update existing BTs to add new knowledge without time consuming batch re-training. The semi-supervised BT utilizes co-training to improve classification accuracy by making use of the information in the unlabeled samples and thus reduce the labeling burden of a human operator. We also proposed compact knowledge representation such that data can be fit into limited memory and a fixed computational complexity can be maintained.