*********************************
There is now a CONTENT FREEZE for Mercury while we switch to a new platform. It began on Friday, March 10 at 6pm and will end on Wednesday, March 15 at noon. No new content can be created during this time, but all material in the system as of the beginning of the freeze will be migrated to the new platform, including users and groups. Functionally the new site is identical to the old one. webteam@gatech.edu
*********************************
Title: Frozen Dictionary based Disease Direction and Dictionary based Novelty Detection for Automated Poultry Monitoring
Committee:
Dr. Anderson, , Advisor
Dr. Davenport, Chair
Dr. Clements
Dr. Daley
Abstract: The objective of the proposed research is to enable detection and classification of anomalies in poultry production facilities through sparse, dictionary-based representations of the auditory environment. This research focuses primarily on the detection of respiratory disease in chickens, but should be extensible to other types of problems or conditions that may arise. The scale of the poultry industry creates significant financial potential for the development of signal processing techniques capable of monitoring varying production settings with minimal human intervention. However, there are numerous challenges to the application of audio signal processing in this domain that must be addressed. For example, the vast majority of the data is unlabeled, and it contains significant amounts of noise. Conditions throughout the day also cause short-term changes in the auditory environment, and the aging of the chickens cause drift over the long-term. The proposed work seeks to overcome these challenges through a frozen dictionary approach to disease detection and several novelty detection approaches based on dictionary learning for sparse representations.
