MEOW - Multi-Objective Evolutionary Weapon Detection

Daniel Dimanov, Colin Singleton, Shahin Rostami, Emili Balaguer-Ballester
Proceedings of the Second International Conference on Automated Machine Learning, PMLR 224:5/1-20, 2023.

Abstract

X-ray screening is crucial for ensuring safety and security in crowded public areas. However, X-ray operators are often overwhelmed by the sheer amount of potential threats to assess; thus, current computer vision-aided systems are designed to alleviate these workloads. In this study, we focus on a key, unresolved challenge for developing such automatic X-ray screening systems: the direct application of existing avant garde computer vision approaches does not necessarily yield satisfactory results in the X-ray medium, hindering the effectiveness of current screening systems. To overcome this drawback, we propose a novel automated machine learning (AutoML) multi-objective approach for neural architecture search (NAS) for concealed weapon detection (MEOW). We benchmark MEOW with the state-of-the-art in two comprehensive scenarios in threat identification: SIXray (a popular, massive X-ray dataset) and Residuals (a proprietary, unpublished dataset provided by our industry partners). MEOW consist of the coalescence of two new components: First, we design a heuristic technique to strongly reduce the high computational cost of neuroevolutionary search while preserving a high performance such that it can be effectively used in real-time industrial settings. Second, we devise a novel ensemble approach for combining multiple discovered architectures simultaneously. Leveraging these two characteristics, MEOW outperforms the state-of-the-art while keeping the NAS overhead to a minimum. More broadly, our results suggest that AutoML has a strong potential for security applications.

Cite this Paper

BibTeX
@InProceedings{pmlr-v224-dimanov23a, title = {MEOW - Multi-Objective Evolutionary Weapon Detection}, author = {Dimanov, Daniel and Singleton, Colin and Rostami, Shahin and Balaguer-Ballester, Emili}, booktitle = {Proceedings of the Second International Conference on Automated Machine Learning}, pages = {5/1--20}, year = {2023}, editor = {Faust, Aleksandra and Garnett, Roman and White, Colin and Hutter, Frank and Gardner, Jacob R.}, volume = {224}, series = {Proceedings of Machine Learning Research}, month = {12--15 Nov}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v224/dimanov23a/dimanov23a.pdf}, url = {https://proceedings.mlr.press/v224/dimanov23a.html}, abstract = {X-ray screening is crucial for ensuring safety and security in crowded public areas. However, X-ray operators are often overwhelmed by the sheer amount of potential threats to assess; thus, current computer vision-aided systems are designed to alleviate these workloads. In this study, we focus on a key, unresolved challenge for developing such automatic X-ray screening systems: the direct application of existing avant garde computer vision approaches does not necessarily yield satisfactory results in the X-ray medium, hindering the effectiveness of current screening systems. To overcome this drawback, we propose a novel automated machine learning (AutoML) multi-objective approach for neural architecture search (NAS) for concealed weapon detection (MEOW). We benchmark MEOW with the state-of-the-art in two comprehensive scenarios in threat identification: SIXray (a popular, massive X-ray dataset) and Residuals (a proprietary, unpublished dataset provided by our industry partners). MEOW consist of the coalescence of two new components: First, we design a heuristic technique to strongly reduce the high computational cost of neuroevolutionary search while preserving a high performance such that it can be effectively used in real-time industrial settings. Second, we devise a novel ensemble approach for combining multiple discovered architectures simultaneously. Leveraging these two characteristics, MEOW outperforms the state-of-the-art while keeping the NAS overhead to a minimum. More broadly, our results suggest that AutoML has a strong potential for security applications.} }
Endnote
%0 Conference Paper %T MEOW - Multi-Objective Evolutionary Weapon Detection %A Daniel Dimanov %A Colin Singleton %A Shahin Rostami %A Emili Balaguer-Ballester %B Proceedings of the Second International Conference on Automated Machine Learning %C Proceedings of Machine Learning Research %D 2023 %E Aleksandra Faust %E Roman Garnett %E Colin White %E Frank Hutter %E Jacob R. Gardner %F pmlr-v224-dimanov23a %I PMLR %P 5/1--20 %U https://proceedings.mlr.press/v224/dimanov23a.html %V 224 %X X-ray screening is crucial for ensuring safety and security in crowded public areas. However, X-ray operators are often overwhelmed by the sheer amount of potential threats to assess; thus, current computer vision-aided systems are designed to alleviate these workloads. In this study, we focus on a key, unresolved challenge for developing such automatic X-ray screening systems: the direct application of existing avant garde computer vision approaches does not necessarily yield satisfactory results in the X-ray medium, hindering the effectiveness of current screening systems. To overcome this drawback, we propose a novel automated machine learning (AutoML) multi-objective approach for neural architecture search (NAS) for concealed weapon detection (MEOW). We benchmark MEOW with the state-of-the-art in two comprehensive scenarios in threat identification: SIXray (a popular, massive X-ray dataset) and Residuals (a proprietary, unpublished dataset provided by our industry partners). MEOW consist of the coalescence of two new components: First, we design a heuristic technique to strongly reduce the high computational cost of neuroevolutionary search while preserving a high performance such that it can be effectively used in real-time industrial settings. Second, we devise a novel ensemble approach for combining multiple discovered architectures simultaneously. Leveraging these two characteristics, MEOW outperforms the state-of-the-art while keeping the NAS overhead to a minimum. More broadly, our results suggest that AutoML has a strong potential for security applications.
APA
Dimanov, D., Singleton, C., Rostami, S. & Balaguer-Ballester, E.. (2023). MEOW - Multi-Objective Evolutionary Weapon Detection. Proceedings of the Second International Conference on Automated Machine Learning, in Proceedings of Machine Learning Research 224:5/1-20 Available from https://proceedings.mlr.press/v224/dimanov23a.html.

Related Material