Related Experiment Video
Updated: Jun 4, 2025
08:47
Author Spotlight: UAV Remote Sensing for Efficient Invasive Plant Biomass Estimation
Published on: February 9, 2024
1.3K
A novel machine learning model for perimeter intrusion detection using intrusion image dataset
Shahneela Pitafi1, Toni Anwar1, I Dewa Made Widia2
1Computer & Information Sciences Department (CISD), Universiti Teknologi PETRONAS, Bandar Seri Iskandar, Perak, Malaysia.
Plos One
|December 19, 2024
Summary
A new machine learning model enhances Perimeter Intrusion Detection Systems (PIDS) for improved accuracy. The developed system offers better detection and classification of perimeter intrusions, boosting security.
Area of Science:
- Computer Science
- Artificial Intelligence
- Security Systems
Background:
- Perimeter Intrusion Detection Systems (PIDS) are vital for physical security.
- Current PIDS face challenges in detection accuracy and activity classification.
Purpose of the Study:
- To develop an improved machine learning model for PIDS.
- To enhance the efficiency and accuracy of intrusion detection and activity classification.
Main Methods:
- Utilized InceptionV3 for feature extraction on intrusion image datasets.
- Applied t-SNE for dimensionality reduction and clustering.
- Enhanced the Density-Based Spatial Clustering of Applications with Noise (DBSCAN) algorithm using Manhattan distance for epsilon estimation.
Main Results:
- The proposed model achieved a silhouette score of 0.86.
- Demonstrated superior performance compared to state-of-the-art techniques.
- Successfully addressed DBSCAN's limitations with high-dimensional and varying density data.
Conclusions:
- The enhanced model significantly improves perimeter intrusion detection accuracy and classification.
- Contributes to enhanced societal security through better perimeter protection.
- Provides a foundation for future research in human activity recognition from image datasets.
