CT Datasets
ALERT leveraged medical CT’s advances to contract with a vendor to obtain representative datasets of packed luggage and reference objects. Luggage was scanned on a state-of-the-art medical CT scanner at the vendor’s factory. Approximately 900 objects were placed in luggage and scanned to produce 62 luggage datasets that span the spectrum of packing, density, arrangement, orientation, and size difficulty.
Section 5.3 of the CT program’s final report contains information on the datasets.
Reconstruction Initiative Dataset
ALERT created a projection and image data database corresponding to scans of objects of interest in the presence of various amounts of clutter using a medical CT scanner. The database was generated using reconstructed images using filtered back-projection that matched the images obtained on the medical scanner.
Section 5.2 of the Reconstruction Initiative’s final report contains information on the datasets.
Automated Threat Recognition (ATR) Initiative
ALERT developed automated target recognition algorithms (ATR) for CT-based explosive detection systems (EDS). The research groups were provided with images from scans of targets packed in bags and scanned on a medical CT scanner. The targets were chosen to create detection scenarios similar to what vendors face when detecting explosives using their CT-based EDS.
Section 5.2 of the ATR initiative’s final report contains information on the datasets.
ALERT Airport Re-Identification Dataset
As part of the ALERT video analytics effort, researchers at Northeastern University and Rensselaer Polytechnic Institute developed an annotated dataset that accurately reflects the real-world person re-identification problem. The dataset was constructed using video data from the six cameras installed post central security checkpoints at an active commercial airport within the United States.