Trainable Structure Tensors for Autonomous Baggage Threat Detection Under Extreme Occlusion

TL;DR

This paper introduces a novel trainable structure tensor framework for instance segmentation that effectively detects contraband items in baggage X-ray scans under extreme occlusion, outperforming existing methods across multiple datasets.

Contribution

The paper presents a new trainable structure tensor approach for instance segmentation that handles occlusion and clutter in baggage X-ray scans, validated on diverse datasets.

Findings

Outperforms state-of-the-art in mean average precision

Validated on four diverse X-ray datasets

Effective on grayscale and colored scans

Abstract

Detecting baggage threats is one of the most difficult tasks, even for expert officers. Many researchers have developed computer-aided screening systems to recognize these threats from the baggage X-ray scans. However, all of these frameworks are limited in identifying the contraband items under extreme occlusion. This paper presents a novel instance segmentation framework that utilizes trainable structure tensors to highlight the contours of the occluded and cluttered contraband items (by scanning multiple predominant orientations), while simultaneously suppressing the irrelevant baggage content. The proposed framework has been extensively tested on four publicly available X-ray datasets where it outperforms the state-of-the-art frameworks in terms of mean average precision scores. Furthermore, to the best of our knowledge, it is the only framework that has been validated on combined grayscale and colored scans obtained from four different types of X-ray scanners.