Online ISSN
1751-9640
Print ISSN
1751-9632
IET Computer Vision
Volume 5, Issue 1, January 2011
Volumes & issues:
Volume 5, Issue 1
January 2011
-
- Author(s): M. Xu ; T. Ellis ; S.J. Godsill ; G.A. Jones
- Source: IET Computer Vision, Volume 5, Issue 1, p. 1 –13
- DOI: 10.1049/iet-cvi.2009.0060
- Type: Article
- + Show details - Hide details
-
p.
1
–13
(13)
A comparative study of four suboptimal tracking algorithms that can cope with missing measurements of low-level visual features during partial occlusion is presented. An approach to identify missing measurements in two-dimensional object tracking is formulated. The comparison starts from a symbolic analysis in Kalman filtering and ends with a performance evaluation in Monte Carlo simulations in which the objects manoeuvre and undergo moderate size variations during occlusion. It is found that the algorithm for estimating unobservables from observables outperforms the others in terms of mean square error, robustness and readiness for implementation. - Author(s): Y.-G. Wu and G.-F. Huang
- Source: IET Computer Vision, Volume 5, Issue 1, p. 14 –22
- DOI: 10.1049/iet-cvi.2008.0061
- Type: Article
- + Show details - Hide details
-
p.
14
–22
(9)
An efficient prediction algorithm for motion vectors in video compression is proposed. Motion estimation is an important part in motion compensated video encoding system, which is used to eliminate the temporal redundancy in video sequence. Motion vectors dominate the quality of decoded video so that the generation of motion vectors is an important topic in video coding. Exhaustive block matching algorithm (EBMA) can get the optimal solution for motion vectors, but it takes much computational burden. In the proposed method, the authors use gray prediction to obtain the motion vectors. Gray prediction can predict the motion vectors quickly and accurately. Several video sequences are used to evaluate the performance in this study, and the experimental results show that the time that the proposed method needed is only 1.93% compared to EBMA and 11.75% compared to three steps searching algorithm whereas the degradation of PSNRY compared to EBMA is about 0.43 dB at most, which is almost invisible to human eyes. - Author(s): J. Lin ; J. Ming ; D. Crookes
- Source: IET Computer Vision, Volume 5, Issue 1, p. 23 –32
- DOI: 10.1049/iet-cvi.2009.0121
- Type: Article
- + Show details - Hide details
-
p.
23
–32
(10)
This study investigates face recognition with partial occlusion, illumination variation and their combination, assuming no prior information about the mismatch, and limited training data for each person. The authors extend their previous posterior union model (PUM) to give a new method capable of dealing with all these problems. PUM is an approach for selecting the optimal local image features for recognition to improve robustness to partial occlusion. The extension is in two stages. First, authors extend PUM from a probability-based formulation to a similarity-based formulation, so that it operates with as little as one single training sample to offer robustness to partial occlusion. Second, they extend this new formulation to make it robust to illumination variation, and to combined illumination variation and partial occlusion, by a novel combination of multicondition relighting and optimal feature selection. To evaluate the new methods, a number of databases with various simulated and realistic occlusion/illumination mismatches have been used. The results have demonstrated the improved robustness of the new methods. - Author(s): G. McGunnigle and J. Dong
- Source: IET Computer Vision, Volume 5, Issue 1, p. 33 –49
- DOI: 10.1049/iet-cvi.2009.0038
- Type: Article
- + Show details - Hide details
-
p.
33
–49
(17)
The authors propose a new photometric stereo (PS) method that augments a conventional four-light approach with coaxial illumination. The new method is based on a four-light, leave-one-out approach: the authors argue that coaxial lighting improves robustness to shadowing and specular highlights. The authors also specify a method to detect pixels that are shadowed or that do not conform to Lambert's law. The approach is tested on cylinders of various materials. The authors found that for dielectrics the method gave reliable results for gradients in the range −2 to 2. The results lead them to conclude that coaxial illumination makes PS more robust to shadow and specularity. - Author(s): X. Yang ; Y. Shi ; B. Yang
- Source: IET Computer Vision, Volume 5, Issue 1, p. 50 –67
- DOI: 10.1049/iet-cvi.2009.0083
- Type: Article
- + Show details - Hide details
-
p.
50
–67
(18)
The authors present a general framework of the construction of biorthogonal wavelets based on Bernstein bases along with theory analysis and application. The presented framework possesses the largest possible regularity, the required vanishing moments and the passband flatness of frequency response of filters. Based on this concept, the authors establish explicit formulas for filters of biorthogonal wavelets with arbitrary odd lengths. Meanwhile, a new family of parametric filters with symmetry is constructed. The choice of filter bank in wavelet compression is a critical issue that affects image quality. In this study, an optimal model of FIR aiming at image compression is brought forward, and the optimal finite impulse response (FIR) filters can be obtained correspondingly through sequential quadratic programming (SQP) and genetic algorithm (GA). The authors demonstrate the performance of the new family of filters given in this study for image compression with very encouraging results. - Author(s): W.L. Li ; Z.P. Yin ; Y.A. Huang ; Y.L. Xiong
- Source: IET Computer Vision, Volume 5, Issue 1, p. 68 –76
- DOI: 10.1049/iet-cvi.2009.0032
- Type: Article
- + Show details - Hide details
-
p.
68
–76
(9)
In this study, the authors propose an iterative registration algorithm with point-based representation. The task is to geometrically transform a mobile object to a stationary object. Without a good initial position for two shapes with high-curvature features, traditional point-based algorithms have many challenges, such as suffering from slow convergence or divergence. For these problems, the authors define a new distance function to describe the point-surface distance, where the point-surface shortest distance is considered. Then, a non-linear optimisation model is established to calculate the optimal transformation. Moreover, the convergence of the proposed algorithm is derived and analysed from the viewpoint of geometrical optimisation. The proposed method does not require surface representation, feature extraction, curvature computation and is more applicable when initial position is rough. Its efficiency and robustness are verified by a series of experiments. - Author(s): Y. Jianchao
- Source: IET Computer Vision, Volume 5, Issue 1, p. 77 –85
- DOI: 10.1049/iet-cvi.2008.0004
- Type: Article
- + Show details - Hide details
-
p.
77
–85
(9)
Video registration in the presence of several segments of non-texture frames remains a challenging issue, although a wide repertoire of image registration algorithms was developed over the last two decades. In this study, the authors proposed an effective causal scheme of video frames registration, and implemented it by the modified two frames intensity match algorithm. In order to achieve its real-time performance, the authors avoid handling a much more complex appearance model in this causal system by using registration parameters as measurement values, which are estimated directly from two consecutive frames via intensity matching. The simple linear and Gaussian system is thus derived and an efficient Kalman filter algorithm can be utilised. As the Kalman filter naturally incorporated the temporal information contained in a video into the estimation of registration parameters, the algorithm developed in this study is quite robust and has good performance even if the processed video contains several segments of non-texture frames.
Visual tracking of partially observable targets with suboptimal filtering
Motion vector generation for video coding by gray prediction
Robust face recognition with partial occlusion, illumination variation and limited training data by optimal feature selection
Augmenting photometric stereo with coaxial illumination
General framework of the construction of biorthogonal wavelets based on Bernstein bases: theory analysis and application in image compression
Three-dimensional point-based shape registration algorithm based on adaptive distance function
Alignment of non-texture video frames using Kalman filter
Most viewed content for this Journal
Article
content/journals/iet-cvi
Journal
5
Most cited content for this Journal
-
Brain tumour classification using two-tier classifier with adaptive segmentation technique
- Author(s): V. Anitha and S. Murugavalli
- Type: Article
-
Driving posture recognition by convolutional neural networks
- Author(s): Chao Yan ; Frans Coenen ; Bailing Zhang
- Type: Article
-
Local directional mask maximum edge patterns for image retrieval and face recognition
- Author(s): Santosh Kumar Vipparthi ; Subrahmanyam Murala ; Anil Balaji Gonde ; Q.M. Jonathan Wu
- Type: Article
-
Fast and accurate algorithm for eye localisation for gaze tracking in low-resolution images
- Author(s): Anjith George and Aurobinda Routray
- Type: Article
-
‘Owl’ and ‘Lizard’: patterns of head pose and eye pose in driver gaze classification
- Author(s): Lex Fridman ; Joonbum Lee ; Bryan Reimer ; Trent Victor
- Type: Article