Abstract
Real-Time Visual Subject Tracking and Classification by Combining Motion Signal Analysis andTridimensional - Shape Feature Classifiers with Group-Induction Boosting Algorithms
Highlights
Many home security products that are available on the market promise to detect intruders at home and notify users via text messages
When a user wants to play back and watch all indoor moving subject activities, he has to watch all of the false-positive parts of the footage as well, wasting countless hours of time
The model involves a boosted learning classifier based on Adaboost, and the innovative use of groupinduction as a method of semi-supervised learning
Summary
Many home security products that are available on the market promise to detect intruders at home and notify users via text messages. These home surveillance platforms often have high rates of false-positives and low tolerance for them. The model involves a boosted learning classifier based on Adaboost, and the innovative use of groupinduction as a method of semi-supervised learning. Some tests and development were performed with semisupervised learning, for the most part of the project, and for our research results, we decided to use a fully-supervised model so as to eliminate any possible noise, mislabelling or bias introduced by the use of semi-supervised learning. The project involved a heat sensor (green) but the results of the heat descriptors are reported in a separate paper [2] (Figure 3).
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: Robotics & Automation Engineering Journal
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
