Accéder directement au contenu Accéder directement à la navigation
Communication dans un congrès

Evaluation Metric for Rate of Background Detection

Abstract : This paper proposes an evaluation metric which derive the effectiveness of background modeling algorithms. Background modeling is a key process on developing visual surveillance systems. The requirement of adapting to dynamic environments has motivated researchers to modify existing background modeling algorithms and develop new algorithms with better adaptability. Having the algorithms developed, credentials of each of the algorithms have to be assessed to exploit their effectiveness. Various evaluation metrics have been used for evaluating the rate of foreground extraction, foreground detection, and overall accuracy. However, the rate of background detection has not been exploited by these metrics. Therefore, this paper would provide an insight to the existing evaluation metrics and introduce our proposed metric for estimating the rate of background detection.
Type de document :
Communication dans un congrès
Liste complète des métadonnées

https://hal-univ-bourgogne.archives-ouvertes.fr/hal-01434938
Contributeur : Le2i - Université de Bourgogne <>
Soumis le : vendredi 13 janvier 2017 - 15:01:34
Dernière modification le : lundi 10 août 2020 - 16:28:02

Identifiants

  • HAL Id : hal-01434938, version 1

Citation

Mohamed Abul Hassan, Malik Aamir Saeed, N.M. Saad, David Fofi. Evaluation Metric for Rate of Background Detection. IEEE International Instrumentation and Measurement Technology Conference (I2MTC), IEEE, May 2016, Taipei, Taiwan. pp.385-389. ⟨hal-01434938⟩

Partager

Métriques

Consultations de la notice

175