Prohibited Items Segmentation via Occlusion-aware Bilayer Modeling

2025-06-13

AI Summary

Addressing two key challenges in X-ray security image instance segmentation—significant appearance disparity between prohibited items and natural objects, and severe occlusion causing mask ambiguity—this paper proposes an occlusion-aware dual-layer mask decoder that explicitly models inter-object occlusion relationships. We introduce PIDray-A and PIXray-A, the first large-scale X-ray instance segmentation datasets featuring fine-grained occlusion annotations. To enable zero-shot transfer, we integrate the Segment Anything Model (SAM) and further refine the decoding process using human-annotated occlusion supervision signals. Evaluated on PIDray-A and PIXray-A, our method achieves an 8.2% improvement in mAP, demonstrating substantially enhanced robustness under heavy occlusion. All code and datasets are publicly released.

Technology Category

• Computer Vision: Segmentation
• Humans and AI: Explainable AI (XAI) for Human Understanding
• Application Domains: Security

Application Category

• Security and Privacy: Data transparency and provenance
• Semantics and Knowledge: Provenance, trust, security and privacy, and ethical issues in managing semantic data
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

Instance segmentation of prohibited items in security X-ray images is a critical yet challenging task. This is mainly caused by the significant appearance gap between prohibited items in X-ray images and natural objects, as well as the severe overlapping among objects in X-ray images. To address these issues, we propose an occlusion-aware instance segmentation pipeline designed to identify prohibited items in X-ray images. Specifically, to bridge the representation gap, we integrate the Segment Anything Model (SAM) into our pipeline, taking advantage of its rich priors and zero-shot generalization capabilities. To address the overlap between prohibited items, we design an occlusion-aware bilayer mask decoder module that explicitly models the occlusion relationships. To supervise occlusion estimation, we manually annotated occlusion areas of prohibited items in two large-scale X-ray image segmentation datasets, PIDray and PIXray. We then reorganized these additional annotations together with the original information as two occlusion-annotated datasets, PIDray-A and PIXray-A. Extensive experimental results on these occlusion-annotated datasets demonstrate the effectiveness of our proposed method. The datasets and codes are available at: https://github.com/Ryh1218/Occ

Problem

• Segment prohibited items in X-ray images despite occlusion
• Bridge appearance gap between X-ray and natural objects
• Model occlusion relationships for accurate instance segmentation

Innovation

• Integrates Segment Anything Model for representation
• Uses occlusion-aware bilayer mask decoder
• Creates occlusion-annotated datasets PIDray-A PIXray-A