Project Description :
Endangered Wildlife are protected in remote land where people are restricted to enter. But intrusions of poachers and illegal loggers still occur due to lack of surveillance to cover huge amount of land. The current usage of stealth ability of the camera is low due to limitations of camera angle of view. Maintenance such as changing batteries and memory cards were troublesome reported by Wildlife Conservative Society (WCS) Malaysia. Remote location with no cellular network access would be difficult to transmit video data. Rangers need a system to react to intrusion on time. Thus, this project aims to develop an audio events recognition for intrusion detection based on vehicle engine, wildlife environmental noise and chainsaw activities. Recent studies has shown that random forest technique has been the best among numerous other technique in prediction of linear predictive coding features for acoustic events prediction with very high accuracy. Random Forest classification and feature extraction of Linear Predictive Coding were employed. Training and testing datasets used were obtained from WCS Malaysia. The findings demonstrate that the accuracy rates achieve up to 86% for indicating an intrusion via audio recognition. This project will be beneficial for wildlife protection agencies in maintaining security as it is less power consuming than the current camera trapping surveillance technique.
Research/Project Team :
- Dr Marina Yusoff ( Project Leader )
- Muhamad Amirul Sadikin Bin Md Afendi
Contact Person :
Dr Marina Yusoff ( marinay@tmsk.uitm.edu.my )