Searching for High Density Material in Cargo Containers Using Gravity Gradiometry

Date published

2017-02-06 09:34

Free to read from

Supervisor/s

Journal Title

Journal ISSN

Volume Title

Publisher

Cranfield University

Department

Type

Poster

ISSN

Format

Citation

Leahy, David (2017). Searching for High Density Material in Cargo Containers Using Gravity Gradiometry. Cranfield Online Research Data (CORD). Poster. https://doi.org/10.17862/cranfield.rd.4597909.v1

Abstract

Poster presented at the 2016 Defence and Security Doctoral Symposium.Imaging cargo containers at ports in this country is an important task, especially if they are being checked for possible fissile material. Some detection methods cannot be used as they are too destructive. A possible alternative is the use of gravity gradiometry - a non-destructive sensing technique which can provide better resolution than straightforward gravity readings, with the trade-off being less penetration power. The resulting inverse problem becomes an underdetermined system of linear equations. This poster looks at applying both a level set method and a genetic algorithm. In its simplest form a level set method uses a level set function to define two distinct regions based on the sign of the function at each point, the boundary being where the function is zero. It then uses a gradient-based iterative method to allow the shape to deform (including splitting and merging) to better fit the data. I explore the use of the colour level set method, which uses more than one level set function to describe many domains. Genetic algorithms are methods which draw inspiration from the process of natural selection using steps such as crossover and mutation. By limiting the population size and by use of a reparameterisation the algorithm can work at the speed required for the time-constraints we have. Both methods have their strengths and weaknesses when applied to this real-life problem. Crown copyright.

Description

Software Description

Software Language

Github

Keywords

'DSDS16 poster', 'DSDS16', 'Genetic', 'Level-set', 'Gravity', 'Condensed Matter Imaging', 'Electrical and Electronic Engineering not elsewhere classified'

DOI

10.17862/cranfield.rd.4597909.v1

Rights

CC BY-NC 4.0

Relationships

Relationships

Supplements

Funder/s

Collections