Joint Sub-component Level Segmentation and Classification for Anomaly Detection within Dual-Energy X-Ray Security Imagery
This addresses the critical need for enhanced transport security by improving anomaly detection in complex electronics items, representing a strong specific gain in a domain-specific application.
The paper tackled the problem of automatically detecting concealed anomalies in cluttered X-ray baggage security imagery by introducing a joint sub-component level segmentation and classification strategy using deep convolutional neural networks, achieving approximately 99% true positive and 5% false positive rates.
X-ray baggage security screening is in widespread use and crucial to maintaining transport security for threat/anomaly detection tasks. The automatic detection of anomaly, which is concealed within cluttered and complex electronics/electrical items, using 2D X-ray imagery is of primary interest in recent years. We address this task by introducing joint object sub-component level segmentation and classification strategy using deep Convolution Neural Network architecture. The performance is evaluated over a dataset of cluttered X-ray baggage security imagery, consisting of consumer electrical and electronics items using variants of dual-energy X-ray imagery (pseudo-colour, high, low, and effective-Z). The proposed joint sub-component level segmentation and classification approach achieve ~99% true positive and ~5% false positive for anomaly detection task.