Doctoral Dissertations

Date of Award

5-2021

Degree Type

Dissertation

Degree Name

Doctor of Philosophy

Major

Nuclear Engineering

Major Professor

Jason P. Hayward

Committee Members

Jason P. Hayward, Hairong Qi, Carolyn Seifert, Peter Angelo

Abstract

The capability of assigning source carrier probabilities by fusing both video and radiation data is demonstrated. Potential targets are automatically detected and tracked with computer vision techniques. Meanwhile, a newly developed method for exploiting the angular response of closely-packed radiation detector arrays points towards a radiation source using short, periodic measurements on the order of 1 s. Source carrier probabilities are then assigned after the object's movement measured from the video data stream is correlated with the radiation directionality estimates over a set period of time. In demonstrating source carrier probability assigned to moving sources, three main contributions were made. A radiation directionality algorithm was developed and characterized for a detector array as a function of unattenuated gamma fluence. The Mean Absolute Error (MAE) was measured to be 8.2 degrees using a 113 uCi Cs-137 source and 17.8 degrees using a 35 uCi Co-60 source, where both sources were placed 172.4 cm away and 1 s measurement times were used. The source localization capabilities of the WIND backpack detector array was characterized for source offsets up to 5.2 m for a variety of experimental variables such as integration time and detector speed. Finally, the data fusion strategy was developed such that probabilities calculated with the directionality algorithm could be appropriately accounted and attributed to tracked targets from the video data stream. It was found that the data fused result improved the radiation only result from a MAE of 18 degrees to 8 degrees for a 113 uCi Cs-137 source at an average distance and speed of 4 m and 0.6 m/s, respectively.

Files over 3MB may be slow to open. For best results, right-click and select "save as..."

Share

COinS