Inspection of commerce entering the country is currently a burdensome process, assisted greatly by non-intrusive inspection (NII) technologies such as X-Ray imaging. There are a variety of singular ML techniques that assist in the adjudication process, many are well known, common and widely available with proper data. However, the challenge is to develop a comprehensive system exhibiting certain properties that leverages many techniques simultaneously and combines results to improve overall performance. Here we describe the concepts, components, data and implementation of a system we use for NII adjudication that is well suited to adapting to future needs. We demonstrate several components of this system the benefits of it. Specifically, we show the combination of results from common methods achieves superior performance in concert. First, ensembeling has proven successful in combining results from a variety of object detection models. We show the impact of using our novel graph clique-based method to increase performance over independent models. Additionally, because of performant and reliable object detection results, we demonstrate how constructing an image level contextualization of a scene allows analysis at the metadata level. Lastly, we leverage variational autoencoder (VAE) based similarity models to describe anomalies. We deploy these algorithms in a hierarchical fashion, developing optical signatures for whole NII scans as well as their components (e.g, entire vehicle, trunk, wheels). Here, we quantify the impacts of the inclusion of these methods in image anomaly detection task and demonstrate them in the extensible framework we have developed for NII applications. This paper will serve as an overview of our approach and show results from both combinatorial techniques.
Inspection of commerce entering the country is currently an onerous, custom and extremely labor-intensive process. The imaging of commerce with 2D x-rays aids the ability to interdict certain items entering the country, reducing risk and increasing national security. We use a variety of ML techniques together to develop automated threat recognition (ATR) algorithms to assist this process. The challenge is to develop a system that incorporates several approaches simultaneously and ensemble results to improve overall performance. We employ several algorithmic techniques to solve the problem. Generally, we have synthetic data generation techniques that can rapidly spin up datasets that are then employed to train ML models. GANs are used to refine the synthetic data to resemble reality more faithfully; thresholding objectively improves top line S/N; semi-supervised methods are used to handle data sparsity and leverage any available unlabeled stream of commerce (SOC) data. Suites of object detectors infer in parallel using common techniques and results are then ensembled using a novel graph-based method. Similarity and change detection along with topological data analysis may leverage historical data to investigate anomalies. This paper is an overview of our approach, generating synthetic data and attempting to unite these ML methods to provide the highest performance for ATR of 2D x-ray data.
Large scale non-intrusive inspection (NII) of commercial vehicles is being adopted in the U.S. at a pace and scale that will result in a commensurate growth in adjudication burdens at land ports of entry. The use of computer vision and machine learning models to augment human operator capabilities is critical in this sector to ensure the flow of commerce and to maintain efficient and reliable security operations. The development of models for this scale and speed requires novel approaches to object detection and novel adjudication pipelines. Here we propose a notional combination of existing object detection tools using a novel ensembling framework to demonstrate the potential for hierarchical and recursive operations. Further, we explore the combination of object detection with image similarity as an adjacent capability to provide post-hoc oversight to the detection framework. The experiments described herein, while notional and intended for illustrative purposes, demonstrate that the judicious combination of diverse algorithms can result in a resilient workflow for the NII environment.
The growing x-ray detection burden for vehicles at Ports of Entry in the US requires the development of efficient and reliable algorithms to assist human operator in detecting contraband. Developing algorithms for large-scale non-intrusive inspection (NII) that both meet operational performance requirements and are extensible for use in an evolving environment requires large volumes and varieties of training data, yet collecting and labeling data for these enivornments is prohibitively costly and time consuming. Given these, generating synthetic data to augment algorithm training has been a focus of recent research. Here we discuss the use of synthetic imagery in an object detection framework, and describe a simulation based approach to determining domain-informed threat image projection (TIP) augmentation.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.