Accession Number : AD1051300


Title :   Multi-Source Fusion for Explosive Hazard Detection in Forward Looking Sensors


Descriptive Note : Technical Report,10 Mar 2014,09 Mar 2017


Corporate Author : MISSISSIPPI STATE UNIV MISSISSIPPI STATE MISSISSIPPI STATE United States


Personal Author(s) : Anderson, Derek


Full Text : https://apps.dtic.mil/dtic/tr/fulltext/u2/1051300.pdf


Report Date : 01 Dec 2016


Pagination or Media Count : 11


Abstract : This proposal is in response to sections 5.2 and 4.3 in the U.S. Army BAA W911NF-12-R-0012. The aim of this research is to improve the U.S. Army's ability to detect landmines and explosive hazards in different scenarios using multiple forward looking (FL) sensors, namely infrared (IR), forward looking ground penetrating radar (FLGPR) and visual spectrum (aka color). This is a real problem that has direct impact on the mobility of the U.S. Army and on the safety of our troops. Scientific advancements will come in the form of novel signal and image processing, data fusion and discrimination (pattern recognition) algorithms for multi-CPU and graphics processor unit (GPU) hardware to autonomously process data from different sensors on various platforms. This research is supported by the U.S. Army Night Vision and Electronic Sensors Directorate (NVESD) countermine division interms of sensors, platforms, data collection and discussions regarding project findings if/when appropriate.


Descriptors :   artificial intelligence software , computer vision , dimensionality reduction , images , probability distributions , frequency domain , pattern recognition , algorithms , detectors , detection , artificial intelligence , feature extraction , signal processing , supervised machine learning , image processing , machine learning


Subject Categories : Cybernetics
      Miscellaneous Detection and Detectors


Distribution Statement : APPROVED FOR PUBLIC RELEASE