Automated Image Processing for X-ray Based Security Scanners
F3-A4 (Phase 1)

Download the 2013 Project Report

X-ray imaging systems relying on backscattered or transmitted energy are widely used in security screening of both individuals and vehicles. In collaboration with an industrial partner, we have focused on developing novel algorithms to address image processing challenges that arise in airport X-ray backscatter (XBS) systems, building on our prior work. Because these systems use very low X-ray doses in order to minimize the radiation exposure of the traveling public, the resulting images have relatively low signal-to-noise ratio (SNR). This creates a need for effective image denoising approaches that, unlike standard smoothing kernels, preserve edge information to allow for the accurate identification of threat objects. This year, we have extended our previous work and have developed new methods for combining multiple denoising estimates to reduce noise while enhancing edge contrast. These methods, demonstrated in this report and elsewhere, have the potential to improve the sensitivity and specificity of automated XBS screening algorithms.

As an offshoot of our work, we published a paper describing the first use of NLM for denoising biomedical signals, showing superior performance to some state-of-the-art wavelet approaches for ECG signals.
ALERT Year 5 Annual Report, Major Contributions
Project Leader
  • Eric Miller
    Tufts University

Faculty and Staff Currently Involved in Project
  • Brian Tracey
    Research Assistant Professor
    Tufts University