Multispectral Processing of Side Looking Synthetic Aperture Acoustic Data for Explosive Hazard Detection
Saved in:
| Title: | Multispectral Processing of Side Looking Synthetic Aperture Acoustic Data for Explosive Hazard Detection |
|---|---|
| Authors: | Murray, Bryce J |
| Source: | Theses and Dissertations |
| Publisher Information: | Scholars Junction |
| Publication Year: | 2018 |
| Subject Terms: | deep learning, convolutional neural network, multispectral, explosive hazard detection, synthetic apeture acoustics |
| Description: | Substantial interest resides in identifying sensors, algorithms and fusion theories to detect explosive hazards. This is a significant research effort because it impacts the safety and lives of civilians and soldiers alike. However, a challenging aspect of this field is we are not in conflict with the threats (objects) per se. Instead, we are dealing with people and their changing strategies and preferred method of delivery. Herein, I investigate one method of threat delivery, side attack explosive ballistics (SAEB). In particular, I explore a vehicle-mounted synthetic aperture acoustic (SAA) platform. First, a wide band SAA signal is decomposed into a higher spectral resolution signal. Next, different multi/hyperspectral signal processing techniques are explored for manual band analysis and selection. Last, a convolutional neural network (CNN) is used for filter (e.g., enhancement and/or feature) learning and classification relative to the full signal versus different subbands. Performance is assessed in the context of receiver operating characteristic (ROC) curves on data from a U.S. Army test site that contains multiple target and clutter types, levels of concealment and times of day. Preliminary results indicate that a machine learned CNN solution can achieve better performance than our previously established human engineered Fraz feature with kernel support vector machine classification. |
| Document Type: | text |
| File Description: | application/pdf |
| Language: | unknown |
| Relation: | https://scholarsjunction.msstate.edu/td/3250; https://scholarsjunction.msstate.edu/context/td/article/4249/viewcontent/etd_01132018_113703.pdf |
| Availability: | https://scholarsjunction.msstate.edu/td/3250 https://scholarsjunction.msstate.edu/context/td/article/4249/viewcontent/etd_01132018_113703.pdf |
| Accession Number: | edsbas.BADA8D2A |
| Database: | BASE |
Nájsť tento článok vo Web of Science | Abstract: | Substantial interest resides in identifying sensors, algorithms and fusion theories to detect explosive hazards. This is a significant research effort because it impacts the safety and lives of civilians and soldiers alike. However, a challenging aspect of this field is we are not in conflict with the threats (objects) per se. Instead, we are dealing with people and their changing strategies and preferred method of delivery. Herein, I investigate one method of threat delivery, side attack explosive ballistics (SAEB). In particular, I explore a vehicle-mounted synthetic aperture acoustic (SAA) platform. First, a wide band SAA signal is decomposed into a higher spectral resolution signal. Next, different multi/hyperspectral signal processing techniques are explored for manual band analysis and selection. Last, a convolutional neural network (CNN) is used for filter (e.g., enhancement and/or feature) learning and classification relative to the full signal versus different subbands. Performance is assessed in the context of receiver operating characteristic (ROC) curves on data from a U.S. Army test site that contains multiple target and clutter types, levels of concealment and times of day. Preliminary results indicate that a machine learned CNN solution can achieve better performance than our previously established human engineered Fraz feature with kernel support vector machine classification. |
|---|