Thesis Defense Announcement for Blake Brockner
June 11, 2019
Faculty and Students,
You are cordially invited to my thesis defense.
Title: Side-attack explosive hazard detection in voxel-space radar using signal processing and convolutional neural networks
When: Wednesday, June 19th at 3:00 p.m.
Where: Simrall Hall, Room 228 (Conference Room)
Candidate: Blake Brockner
Degree: Masters, Electrical and Computer Engineering
Committee:
Dr. John Ball
(Major Professor )
Dr. Derek Anderson
(Committee Member)
Dr. Christopher Archibald
(Committee Member)
Abstract:
The detection of concealed explosive hazards is an important task for both current and former conflict zones. Being capable of detecting these devices prior to activation saves both military and civilian lives. There are many current methods of detection, but most require being dangerously close or above the targets to detect them. In my research I explore the use high-resolution near field 3D synthetic aperture radar (SAR) imagery in the detection of explosive hazards. I developed a multi-stage detection algorithm for locating anomalies within the 3D image and making a probability-based classification. I then modified the algorithm by replacing the classification portion with a convolutional neural network. The results for both methods were measured against the detection capabilities of an experienced human for the same datasets.