期刊
IEEE TRANSACTIONS ON SYSTEMS MAN CYBERNETICS-SYSTEMS
卷 52, 期 11, 页码 6937-6951出版社
IEEE-INST ELECTRICAL ELECTRONICS ENGINEERS INC
DOI: 10.1109/TSMC.2021.3131421
关键词
X-ray imaging; Semantics; Training; Image segmentation; Deep learning; Task analysis; Detectors; Baggage X-ray scans; incremental learning; instance segmentation; semantic segmentation
资金
- Khalifa University [CIRA-2019-047]
- Abu Dhabi Department of Education and Knowledge (ADEK) [AARE19-156]
This study introduces a novel method for identifying contraband items in baggage X-ray scans, utilizing an encoder-decoder network structure that evolves during training to recognize individual instances, and shows superior performance in challenging scenarios compared to current state-of-the-art methods.
Screening cluttered and occluded contraband items from baggage X-ray scans is a cumbersome task even for the expert security staff. This article presents a novel strategy that extends a conventional encoder-decoder architecture to perform instance-aware segmentation and extract merged instances of contraband items without using any additional subnetwork or an object detector. The encoder-decoder network first performs conventional semantic segmentation and retrieves cluttered baggage items. The model then incrementally evolves during training to recognize individual instances using significantly reduced training batches. To avoid catastrophic forgetting, a novel objective function minimizes the network loss in each iteration by retaining the previously acquired knowledge while learning new class representations and resolving their complex structural interdependencies through Bayesian inference. A thorough evaluation of our framework on two publicly available X-ray datasets shows that it outperforms state-of-the-art methods, especially within the challenging cluttered scenarios, while achieving an optimal tradeoff between detection accuracy and efficiency.
作者
我是这篇论文的作者
点击您的名字以认领此论文并将其添加到您的个人资料中。
推荐
暂无数据