CT-Based Explosive Detection Equipment: Improved Reconstruction and Accelerated Deployment
ADSA07

The final report for this workshop is available at:

https://alert.northeastern.edu/transitioning-technology/adsa/final-reports-and-presentations/

ADSA07 focused on reconstruction algorithms for CT-based explosion detection systems. This workshop was the seventh in a series dealing with algorithm development for security applications.

The topic of reconstruction was chosen for the workshop in order to support the Department of Homeland Security’s (DHS) objective of improving the detection performance of existing technologies. Detection performance is defined as increased probability of detection, decreased probability of false alarms, lower threat mass and an increased number of types of explosives.

The key topics that were addressed at the workshop are as follows:

  • CT reconstruction for few and many-view scanners
  • Pre-processing, post-processing, and model-based methods for artifact reduction
  • Advances in segmentation algorithms for CT-based explosive detection scanners
  • Tools for simulating explosive detection equipment
  • Accelerating the deployment of advances from 3rd parties
  • Review of DHS product acceptance testing and TSA deployment processes

The workshop was successful in the sense that it fostered interaction between third parties and vendors, reducing barriers to their working together, now and in the future.  It also directly led to increased third party involvement in the development of advanced reconstruction algorithms. This conclusion is based on anecdotal evidence of the number of third parties engaging in discussions with vendors during the workshop and the editors’ knowledge of third parties consulting for the vendors.

Workshop Outcomes

  • There are improved reconstruction algorithms available for CT-based explosive detection equipment. In particular, these algorithms may reduce artifacts such as streaks and cupping. Such improved algorithms may lead to improved explosive detection performance. Algorithms, capabilities, characteristics and features that were highlighted as having the potential to provide such gains include:
    • Iterative reconstruction techniques, which are also known as model-based and statistical reconstruction.
    • Improved filtered back-projection.
    • Sinogram processing.
    • Algorithms targeted to reduce CT artifacts, especially artifacts caused by metal, beam hardening and scatter.
    • Reconstruction algorithms that perform dual-energy decomposition simultaneously with reconstruction.
    • Algorithms that perform reconstruction and segmentation simultaneously.
    • Algorithms that exploit prior information, learning, and compressive sensing.

The following infrastructure should be put in place in order to facilitate and accelerate the development of improved reconstruction algorithms.

  • Public domain computer simulations of security CT scanners, along with the development of standardized simulated objects and simulated packing algorithms.
  • Relevant metrics of image quality instead of actually measuring detection performance. At present there is no precedent for using image quality metrics to assess the performance of CT-based explosive detection equipment.
  • Projections and meta-data corresponding to scans of standard test objects on a CT scanner.
  • Problem statements describing problems that are of interest to the field that are not classified or sensitive security information.
  • Funding for researchers from DHS, TSA and industry.
  • Incentives from the TSA for vendors to deploy equipment with improved detection performance. These incentives will lead to the deployment of advanced reconstruction algorithms.