This paper covers advances made in the trip- wire detection aspect of the Landmine Detection 4.0 project. This includes algorithmic improvements - most importantly a comparison between a gradient-sigmoid filter and a derivative subtraction filter - and testing of these algorithms on tripwire scenes. We used a Canon DSLR camera for testing, which provides higher resolution and a different spectral range than the FLIR thermal camera we had used in previous testing. We have also begun investigating the use of near-infrared cameras as a possible alternative. Additionally, we have investigated methods to reduce false-positive detections within images with the intention to implement and test some of these methods in the near future.
Sensors and Algorithm Evaluation for Tripwire Detection in the Landmine Detection 4.0 Project / Jonathon Sinton, Timothy D. Bechtel, Fronefield Crawford, Luca Bossi, Lorenzo Capineri, Pierluigi Falorni, Gabriella Sallai, Anastasia Kuske. - ELETTRONICO. - (2020), pp. 0-0. (Intervento presentato al convegno 2020 IEEE International Workshop on Metrology for Industry 4.0 & IoT) [10.1109/MetroInd4.0IoT48571.2020.9138195].
Sensors and Algorithm Evaluation for Tripwire Detection in the Landmine Detection 4.0 Project
Luca Bossi;Lorenzo Capineri;Pierluigi Falorni;
2020
Abstract
This paper covers advances made in the trip- wire detection aspect of the Landmine Detection 4.0 project. This includes algorithmic improvements - most importantly a comparison between a gradient-sigmoid filter and a derivative subtraction filter - and testing of these algorithms on tripwire scenes. We used a Canon DSLR camera for testing, which provides higher resolution and a different spectral range than the FLIR thermal camera we had used in previous testing. We have also begun investigating the use of near-infrared cameras as a possible alternative. Additionally, we have investigated methods to reduce false-positive detections within images with the intention to implement and test some of these methods in the near future.File | Dimensione | Formato | |
---|---|---|---|
Sensors and Algorithm Evaluation for Tripwire Detection in the Landmine Detection 4.0 Project.pdf
Accesso chiuso
Tipologia:
Pdf editoriale (Version of record)
Licenza:
Tutti i diritti riservati
Dimensione
3.57 MB
Formato
Adobe PDF
|
3.57 MB | Adobe PDF | Richiedi una copia |
I documenti in FLORE sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.