A performance analysis of dense stereo correspondence algorithms and error reduction techniques
- Khan, N.A., Robinson, P.E., Nel, A.L.
- Authors: Khan, N.A. , Robinson, P.E. , Nel, A.L.
- Date: 2017
- Subjects: Stereo vision , Dense stereo correspondence algorithms , Dense stereo evaluation
- Language: English
- Type: Conference proceedings
- Identifier: http://hdl.handle.net/10210/260015 , uj:27369 , Citation: Khan, N.A., Robinson, P.E. & Nel, A.L. 2017. A performance analysis of dense stereo correspondence algorithms and error reduction techniques.
- Description: Abstract: Dense stereo correspondence has been intensely studied and there exists a wide variety of proposed solutions in the literature. Different datasets have been constructed to test stereo algorithms, however, their ground truth formation and scene types vary. In this paper, state-of-the-art algorithms are compared using a number of datasets captured under varied conditions, with accuracy and density metrics forming the basis of a performance evaluation. Pre- and post-processing disparity map error reduction techniques are quantified.
- Full Text:
- Authors: Khan, N.A. , Robinson, P.E. , Nel, A.L.
- Date: 2017
- Subjects: Stereo vision , Dense stereo correspondence algorithms , Dense stereo evaluation
- Language: English
- Type: Conference proceedings
- Identifier: http://hdl.handle.net/10210/260015 , uj:27369 , Citation: Khan, N.A., Robinson, P.E. & Nel, A.L. 2017. A performance analysis of dense stereo correspondence algorithms and error reduction techniques.
- Description: Abstract: Dense stereo correspondence has been intensely studied and there exists a wide variety of proposed solutions in the literature. Different datasets have been constructed to test stereo algorithms, however, their ground truth formation and scene types vary. In this paper, state-of-the-art algorithms are compared using a number of datasets captured under varied conditions, with accuracy and density metrics forming the basis of a performance evaluation. Pre- and post-processing disparity map error reduction techniques are quantified.
- Full Text:
Comparison of background subtraction techniques under sudden illumination changes
- Nel, A.L., Robinson, P.E., Reyneke, C.J.F.
- Authors: Nel, A.L. , Robinson, P.E. , Reyneke, C.J.F.
- Date: 2014
- Subjects: Background subtraction , Illumination changes
- Type: Article
- Identifier: uj:5043 , ISBN ISBN 978-0-620-62617-0. , http://hdl.handle.net/10210/13563
- Description: This paper investigates three background modelling techniques that have potential to be robust against sudden and gradual illumination changes for a single, stationary camera. The first makes use of a modified local binary pattern that considers both spatial texture and colour information. The second uses a combination of a frame-based Gaussianity Test and a pixel-based Shading Model to handle sudden illumination changes. The third solution is an extension of a popular kernel density estimation (KDE) technique from the temporal to spatio-temporal domain using 9-dimensional data points instead of pixel intensity values and a discrete hyperspherical kernel instead of a Gaussian kernel. A number of experiments were performed to provide a com- parison of these techniques in regard to classfication accuracy.
- Full Text:
- Authors: Nel, A.L. , Robinson, P.E. , Reyneke, C.J.F.
- Date: 2014
- Subjects: Background subtraction , Illumination changes
- Type: Article
- Identifier: uj:5043 , ISBN ISBN 978-0-620-62617-0. , http://hdl.handle.net/10210/13563
- Description: This paper investigates three background modelling techniques that have potential to be robust against sudden and gradual illumination changes for a single, stationary camera. The first makes use of a modified local binary pattern that considers both spatial texture and colour information. The second uses a combination of a frame-based Gaussianity Test and a pixel-based Shading Model to handle sudden illumination changes. The third solution is an extension of a popular kernel density estimation (KDE) technique from the temporal to spatio-temporal domain using 9-dimensional data points instead of pixel intensity values and a discrete hyperspherical kernel instead of a Gaussian kernel. A number of experiments were performed to provide a com- parison of these techniques in regard to classfication accuracy.
- Full Text:
Mean shift object tracking with occlusion handling
- De Villiers, B.Z., Clarke, W.A., Robinson, P.E.
- Authors: De Villiers, B.Z. , Clarke, W.A. , Robinson, P.E.
- Date: 2012
- Subjects: Object tracking , Target objects
- Type: Article
- Identifier: uj:6283 , ISBN 978-0-620-54601-0 , http://hdl.handle.net/10210/9889
- Description: An object tracking algorithm using the Mean Shift framework is presented which is largely invariant to both partial and full occlusions, complex backgrounds and change in scale. Multiple features are used to gain a descriptive representation of the target object. Image moments are used to determine the scale of the target object. A kalman filter is used to successfully track the target object through partial and full occlusions, the Bhattacharyya coefficient is used to determine the measurement noise estimation.
- Full Text:
- Authors: De Villiers, B.Z. , Clarke, W.A. , Robinson, P.E.
- Date: 2012
- Subjects: Object tracking , Target objects
- Type: Article
- Identifier: uj:6283 , ISBN 978-0-620-54601-0 , http://hdl.handle.net/10210/9889
- Description: An object tracking algorithm using the Mean Shift framework is presented which is largely invariant to both partial and full occlusions, complex backgrounds and change in scale. Multiple features are used to gain a descriptive representation of the target object. Image moments are used to determine the scale of the target object. A kalman filter is used to successfully track the target object through partial and full occlusions, the Bhattacharyya coefficient is used to determine the measurement noise estimation.
- Full Text:
Reflex assisted walking for a hexapod robot
- Marais, S.T., Nel, A.L., Robinson, P.E.
- Authors: Marais, S.T. , Nel, A.L. , Robinson, P.E.
- Date: 2016
- Subjects: Hexapod robots
- Language: English
- Type: Conference proceedings
- Identifier: http://hdl.handle.net/10210/217572 , uj:21656 , Citation: Marais, S.T., Nel, A.L. & Robinson, P.E. 2016. Reflex assisted walking for a hexapod robot.
- Description: Abstract: This paper describes the performance of a hexapod robot that uses low level leg reflexes to aid in walking over uneven terrain, and is currently being developed at the University of Johannesburg. The goal of this research is a robot able to deploy in both the inspection and search and rescue roles, within an underground mine environment. The robot has six legs with three degrees of freedom per leg, and is equipped with a two degree of freedom arm with a sensor payload attached to a pan-tilt system. Throughout the development of this new robot the Design Science Research Methodology was used to guide the decision making process. This paper presents an overview of the robot, including the control architecture, and the testing conducted to verify the robot’s performance when walking over a laboratory test field.
- Full Text:
- Authors: Marais, S.T. , Nel, A.L. , Robinson, P.E.
- Date: 2016
- Subjects: Hexapod robots
- Language: English
- Type: Conference proceedings
- Identifier: http://hdl.handle.net/10210/217572 , uj:21656 , Citation: Marais, S.T., Nel, A.L. & Robinson, P.E. 2016. Reflex assisted walking for a hexapod robot.
- Description: Abstract: This paper describes the performance of a hexapod robot that uses low level leg reflexes to aid in walking over uneven terrain, and is currently being developed at the University of Johannesburg. The goal of this research is a robot able to deploy in both the inspection and search and rescue roles, within an underground mine environment. The robot has six legs with three degrees of freedom per leg, and is equipped with a two degree of freedom arm with a sensor payload attached to a pan-tilt system. Throughout the development of this new robot the Design Science Research Methodology was used to guide the decision making process. This paper presents an overview of the robot, including the control architecture, and the testing conducted to verify the robot’s performance when walking over a laboratory test field.
- Full Text:
Robust single image noise estimation from approximate local statistics
- Roodt, Y., Clarke, W., Robinson, P.E., Nel. A.
- Authors: Roodt, Y. , Clarke, W. , Robinson, P.E. , Nel. A.
- Date: 2012
- Language: English
- Identifier: http://hdl.handle.net/10210/16269 , uj:15756 , Citation: Roodt, Y. et al. 2012. Robust single image noise estimation from approximate local statistics. Pattern Recognition Association of South Africa, 2012.
- Description: Abstract: Abstract—A novel method for estimating the variance and standard deviation of the additive white Gaussian noise contained in an image will be presented. Only a single image is used to estimate the noise properties. Local image outliers are discarded, this allows us to separate the additive zero mean white Gaussian noise contained in a noisy image from the original image structure...
- Full Text:
- Authors: Roodt, Y. , Clarke, W. , Robinson, P.E. , Nel. A.
- Date: 2012
- Language: English
- Identifier: http://hdl.handle.net/10210/16269 , uj:15756 , Citation: Roodt, Y. et al. 2012. Robust single image noise estimation from approximate local statistics. Pattern Recognition Association of South Africa, 2012.
- Description: Abstract: Abstract—A novel method for estimating the variance and standard deviation of the additive white Gaussian noise contained in an image will be presented. Only a single image is used to estimate the noise properties. Local image outliers are discarded, this allows us to separate the additive zero mean white Gaussian noise contained in a noisy image from the original image structure...
- Full Text:
- «
- ‹
- 1
- ›
- »