Consiglio Nazionale delle Ricerche

Tipo di prodottoArticolo in rivista
TitoloBackground estimation by weightless neural networks
Anno di pubblicazione2017
Formato
  • Elettronico
  • Cartaceo
Autore/iMassimo De Gregorio and Maurizio Giordano
Affiliazioni autoriIstituto di Scienze Applicate e Sistemi Intelligenti CNR Istituto di Calcolo e Reti ad Alte Prestazioni CNR
Autori CNR e affiliazioni
  • MASSIMO DE GREGORIO
  • MAURIZIO GIORDANO
Lingua/e
  • inglese medio (1100-1500)
AbstractInitial background estimation in video processing serves as bootstrapping model for moving object detection based on background subtraction. In long-term videos, the initial background model may require a constant update as long as the environmental changes happen (slowly or suddenly). In this paper we approach the background initialization together with its constant updating in time by modeling video background as ever-changing states of weightless neural networks. The result is a background estimation method based on a weightless neural network, called BEWiS. The approach proposed in this work is simple: background estimation at each pixel is carried out by weightless neural networks designed to learn pixel color frequency during video play, and all networks share the same rule for memory retention during training. This approach has the advantage of providing a useful background model at the very beginning of the video, since it operates in unsupervised mode. On the other hand, depending on the video scene, the pixel-level learning rule can be tuned to tackle the specificities and difficulties of the scene. The approach presented in this work has been experimented on the public Scene Background Initialization 2015 dataset and on the Scene Background Modeling Contest 2016 dataset, and it showed a performance comparable or superior to state-of-the-art methods.
Lingua abstractinglese
Altro abstract-
Lingua altro abstract-
Pagine da55
Pagine a65
Pagine totali11
RivistaPattern recognition letters
Attiva dal 1982
Editore: North-Holland - Amsterdam
Paese di pubblicazione: Paesi Bassi
Lingua: inglese
ISSN: 0167-8655
Titolo chiave: Pattern recognition letters
Titolo proprio: Pattern recognition letters.
Titolo abbreviato: Pattern recogn. lett.
Numero volume della rivista96
Fascicolo della rivista-
DOI10.1016/j.patrec.2017.05.029
Verificato da refereeSì: Internazionale
Stato della pubblicazionePreprint
Indicizzazione (in banche dati controllate)
  • Scopus (Codice:2-s2.0-85020103756)
  • ISI Web of Science (WOS) (Codice:000411422300007)
Parole chiaveBackground model, Video surveillance, Weightless neural networks
Link (URL, URI)http://www.sciencedirect.com/science/article/pii/S0167865517301927
Titolo parallelo-
Licenza-
Scadenza embargo-
Data di accettazione-
Note/Altre informazioni-
Strutture CNR
  • ICAR — Istituto di calcolo e reti ad alte prestazioni
  • ISASI — Istituto di Scienze Applicate e Sistemi Intelligenti "Eduardo Caianiello"
Moduli/Attività/Sottoprogetti CNR
  • DIT.AD022.055.001 : Intelligenza Computazionale
Progetti Europei-
Allegati
Background estimation by weightless neural networks (documento privato )
Descrizione: Reprint
Tipo documento: application/pdf