On the Impact of Object and Sub-component Level Segmentation Strategies for Supervised Anomaly Detection within X-ray Security Imagery.
Neelanjan BhowmikYona Falinie A. GausSamet AkcayJack W. BarkerToby P. BreckonPublished in: CoRR (2019)
Keyphrases
- x ray
- anomaly detection
- intrusion detection
- medical imaging
- computer security
- network security
- pixel level
- x ray images
- unsupervised learning
- digital x ray images
- malware detection
- intrusion detection system
- network traffic
- detecting anomalies
- poor contrast
- tomographic images
- intraoperative
- ct volume
- network intrusion detection
- anomalous behavior
- one class support vector machines
- ct and mr images
- segmentation algorithm
- three dimensional
- negative selection algorithm
- ct scans
- information security
- transmission electron microscopy
- segmentation method
- d objects
- detect anomalies
- computer vision
- electron microscopy
- medical images
- machine learning
- high resolution
- accurate segmentation
- learning algorithm
- image analysis
- shape prior
- image data
- intrusion prevention
- multi view
- data mining