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 AkçayJack W. BarkerToby P. BreckonPublished in: ICMLA (2019)
Keyphrases
- x ray
- anomaly detection
- intrusion detection
- medical imaging
- computer security
- network security
- pixel level
- x ray images
- unsupervised learning
- anomalous behavior
- intrusion detection system
- detecting anomalies
- intraoperative
- digital x ray images
- network intrusion detection
- network traffic
- tomographic images
- malware detection
- medical images
- d objects
- poor contrast
- ct volume
- transmission electron microscopy
- detect anomalies
- segmentation algorithm
- three dimensional
- image segmentation
- ct scans
- information security
- supervised learning
- segmentation method
- ct and mr images
- learning algorithm
- one class support vector machines
- intrusion prevention
- pre operative
- negative selection algorithm
- automatic segmentation
- electron microscopy
- high resolution
- image data
- object tracking
- machine learning
- feature selection
- accurate segmentation
- image processing
- pattern recognition