Automated X-ray image analysis for cargo security: Critical review and future promise

Area of Science:

• Computer Vision and Image Processing
• Artificial Intelligence in Security

Background:

• Growing global trade necessitates efficient cargo inspection.
• Increasing security threats require advanced methods to detect smuggled items.
• Current manual inspection methods are insufficient for high cargo volumes.

Purpose of the Study:

• To review the current state of automated image analysis for X-ray cargo imagery.
• To identify research gaps and propose future research directions in this immature field.
• To provide a structured overview of existing techniques.

Main Methods:

• Categorization of automated image analysis into preprocessing and understanding.
• Review of techniques including image manipulation, quality improvement, Threat Image Projection (TIP), material discrimination, segmentation, Automated Threat Detection (ATD), and Automated Contents Verification (ACV).
• Exploration of related domains like baggage X-ray analysis for insights.

Main Results:

• The field of automated X-ray cargo image analysis is relatively immature.
• Key components include image preprocessing (manipulation, quality, TIP, segmentation) and image understanding (ATD, ACV).
• Significant gaps exist in the current literature requiring further investigation.

Conclusions:

• Automated analysis of X-ray cargo imagery is crucial for enhancing security and inspection efficiency.
• Further research is needed to address identified literature gaps.
• Leveraging insights from related X-ray imaging domains can accelerate progress.