Adaptive Background Modeling for Effective Ghost Removal and Robust Left Object Detection

2010 ◽  
Author(s):  
Hwiseok Yang ◽  
Yunyoung Nam ◽  
We-Duke Cho ◽  
Yoo-Joo Choi
Keyword(s):  
Object Detection ◽  
Background Modeling ◽  
Left Object

Towards Robust Object Detection: Integrated Background Modeling Based on Spatio-temporal Features

2010 ◽  
pp. 201-212 ◽  
Author(s):  
Tatsuya Tanaka ◽  
Atsushi Shimada ◽  
Rin-ichiro Taniguchi ◽  
Takayoshi Yamashita ◽  
Daisaku Arita
Keyword(s):  
Object Detection ◽  
Background Modeling ◽  
Temporal Features ◽  
Spatio Temporal

scholarly journals Kernel Density Estimation and Correntropy based Background Modeling and Camera Model Parameter Estimation for Underwater Video Object Detection

2021 ◽  
Author(s):  
Susmita Panda ◽  
Pradipta Kumar Nanda
Keyword(s):  
Object Detection ◽  
Density Estimation ◽  
Kernel Density Estimation ◽  
Kernel Density ◽  
Background Modeling ◽  
Model Parameters ◽  
Video Object ◽  
Underwater Video ◽  
Camera Model ◽  
Model Learning

Abstract Underwater video object detection is challenging because of the complex background and the movement of the camera. In order to address this, we propose a novel scheme of simultaneously estimating the camera model parameters and detecting the object. The object detection phase includes background modeling and its learning. Background is modeled by the proposed Spatial Kernel Density Estimation (SKDE) model and the model learning happens in the SKDE feature space. Background modeling and its learning is pixel based approach. The model histograms learn the new pixel through its histogram representation. Our learning and classification strategy is different from the Heikkila et al. [17] in the context of similarity measure. We have proposed the correntropy based similarity measure that is used for model learning and pixel classification. The camera model parameters are estimated by 2D optimization method where we have used the corner features of an object at subpixel accuracy level. These subpixel level features are used in the proposed pipelining framework for model parameters estimation. The estimated model parameters are used to transform the input frame, which in turn is used for model learning and classification. The proposed scheme has been tested with underwater video frames from six data sets. The efficacy of the proposed scheme is compared with seven existing schemes and it is found that the proposed scheme exhibits improved performance as compared to the existing methods.

Sign in / Sign up

Export Citation Format

Share Document