Application of Detection and Recognition Algorithms to Persistent Wide Area Surveillance

Fehlmann, Shannon; Booth, David T.; Janney, Pranam; Pontecorvo, Carmine; Aquilina, Peter Vincent; Scoleri, Tony; Redding, N.J.; Christie, Robert

doi:10.1109/dicta.2013.6691482

Cited by 6 publications

(4 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The observers and filters described in this paper may be suitable for point-target tracking and low-level imageprocessing functions (respectively) within a computer-vision system, for video moving-target indication, on an airborne platform [32], [35], [36]. Fixed-gain trackers are suitable in these systems, for the following reasons: High data-rates (e.g.…”

Section: A Applicationmentioning

confidence: 99%

Fixed-Gain Augmented-State Tracking-Filters

Kennedy¹

2017

Preprint

View full text Add to dashboard Cite

A procedure for the design of fixed-gain tracking filters, using an augmented-state observer with signal and interference subspaces, is proposed. The signal subspace incorporates an integrating Newtonian model and a second-order maneuver model that is matched to a sustained constant-g turn; the deterministic interference model creates a Nyquist null for smoother track estimates. The selected models provide a simple means of shaping and analyzing the (transient and steady-state) response of tracking-filters of elevated order.

show abstract

Section: A Applicationmentioning

confidence: 99%

Fixed-Gain Augmented-State Tracking-Filters

Kennedy¹

2017

Preprint

View full text Add to dashboard Cite

show abstract

“…Elevation angle of the sensor is estimated using the imagery metadata [11]. Imagery metadata supplies the platform's latitude, longitude, altitude, absolute yaw, pitch and roll angles along with the sensor's yaw, pitch and roll angles relative to airborne platform (see Fig.…”

Section: Elevationmentioning

confidence: 99%

“…Solving for the sensor's absolute pitch with reference to a target object involved two 3D transformations, transforming into a north-east-down (NED) frame and then computing from an NED frame. For our work, we require an approximate elevation angle of the sensor relative to the target object; hence a more direct approach as suggested in [11] was implemented. Considering A = [A lat , A long , A alt ] as a vector consisting of latitude, longitude and altitude of the airborne platform's position, F = [F lat , F long , F hgt ] as a vector consisting of latitude, longitude and height above sea level of the point at the centre of the frame and the location on the ground in the direction of the sensor denoted by G = [G lat , G long , F hgt ], a vector consisting of latitude , longitude and height above mean sea level.…”

Section: Elevationmentioning

confidence: 99%

Pose-invariant vehicle identification in aerial electro-optical imagery

Janney

Booth

2015

Machine Vision and Applications

Self Cite

View full text Add to dashboard Cite

In digital airborne electro-optical imagery, the identification of objects, particularly vehicles, has an important role in wide-area search and surveillance applications. We propose an identification and pose estimation approach based on maximising the correlation of features in an image with projections of 3D models. It has been applied to imagery collected in a controlled laboratory environment as well as imagery collected during airborne field trials. The results show good discrimination between different vehicle classes, although performance is degraded by vehicle camouflage and low-resolution imagery. Our approach is scalable, in terms of database size and feature sets, and computationally efficient.

show abstract

“…This GPU has 960 CUDA cores but running only at 980 MHz due to the new efficiency goal of this architecture [114]. CUDA platform has been extensively used in many research fields such as computational finance [115,116], astronomy related [98], medical image [99,100], simulation of clustering process of protein structures [117], parallel data mining [118] and defence-related applications [119,120]. Particularly in this thesis, the power of Nvidia GPUs has been used for accelerating the simulation of wet and dry etching processes applied to micromachining of microstructures [19,20,102].…”

Section: Gpu Features Tesla Keplermentioning

confidence: 99%

Study, Modelling and Implementation of the Level Set Method Used in Micromachining Processes

Álvaro¹

View full text Add to dashboard Cite

"Sí com lo taur se'n va fuyt pel desert quantés sobrat per son semblant qui·l força, ne torna may fins ha cobrada força per destruir aquell qui l'ha desert: tot enaxí·m cové lunyar de vós, car vostre gest mon esforç ha confús: no tornaré fins del tot haja fus la gran pahor qui·m toll ser delitós." Ausiàs March, 1397 - †1459 Per a Marc AgradecimientosSin duda, todo el trabajo que hay detrás de esta tesis ha sido posible gracias a la colaboración y dedicación de muchos. Todos ellos merecen ser agradecidos por haber contribuido a su manera en la culminación de este trabajo.Agradecer sinceramente a mis directores y tutores, Ximo Cerdà y Ricardo Colom, por permitir formarme como investigador y ofrecerme sus consejos. Ellos me han tratado con respeto y nunca han puesto freno a mis ambiciones, más bien las han engrandecido. Gracias por guiar toda esta aventura.Una mención muy especial se merecen MiguelÁngel Gosálvez y Néstor Ferrando. Les tengo que agradecer la gran suerte que he tenido de poder colaborar con dos eminencias como ellos. Sus grandes conocimientos han permitido obtener los mejores resultados de este trabajo, además de haber sido una experiencia gratificante. Gracias por su constante apoyo y su ayuda infinita.Asimismo, dar las gracias también a Vicente Herrero, Jose María Monzó, Ana Ros y Ramón J. Aliaga. Ellos dieron vida a un ambiente de trabajo idóneo, aportándome mucho más que ayuda y conocimientos. Del mismo modoÁngel Tébar, Rafael Gadea y Jorge Daniel siempre han estado dispuestos a ayudar en todo cuanto han podido.Mi estancia en el instituto Fraunhofer de Erlangen me ha permitido acelerar mi formación y vivir unas experiencias de lo más satisfactorias. Todo ello ha sido gracias a la magnífica gente con la que he tenido el placer de colaborar y convivir. Gracias a Eberhard Bär y a toda la incontable gente del Fraunhofer IISB.Un mérito especial se merecen mi familia y mis amigos por aguantarme todo este tiempo sin perder la paciencia conmigo ni en los peores momentos. Es de un valor incalculable tener tantos hombros en los que poder dejarte caer para coger las fuerzas suficientes para seguir. Gracias Isidor y Lourdes, gracias Ferran y Andreu, y gracias Vicente, Enric y David.iii No podía faltar mi particular aimia, Nuria.Ella se merece un eterno agradecimiento por haber tenido que vivir todo esto conmigo. Compañera de todas y cada una de las experiencias, buenas y malas, de esta larga aventura. Animándome en todo momento, siempre ayudándome con un altruismo deslumbrante y enseñándome una inmensidad de conocimientos. Ella, ha sido y es, mi llir entre cards.Una vez más, muchísimas gracias a todos. iv SummaryThe main topic of the present thesis is the improvement of fabrication processes simulation by means of the Level Set (LS) method. The LS is a mathematical approach used for evolving fronts according to a motion defined by certain laws. The main advantage of this method is that the front is embedded inside a higher dimensional function such that updating this function instead of directly the front it...

show abstract

Application of Detection and Recognition Algorithms to Persistent Wide Area Surveillance

Cited by 6 publications

References 15 publications

Fixed-Gain Augmented-State Tracking-Filters

Fixed-Gain Augmented-State Tracking-Filters

Pose-invariant vehicle identification in aerial electro-optical imagery

Study, Modelling and Implementation of the Level Set Method Used in Micromachining Processes

Contact Info

Product

Resources

About