Eam Khwang Teoh

Analysis of gray level corner detection

Abstract In this paper the analysis of gray level corner detection has been carried out. Performances of various cornerness measures are discussed with respect to four performances of robustness: detection, localization, stability and complexity. We have analyzed the interior differential features of the image surface of these cornerness measures. This paper presents a new approach called gradient-direction corner detector for the corner detection which is developed from the popular Plessey corner detection. The gradient-direction corner detector is based on the measure of the gradient module of the image gradient direction and the constraints of the false corner response suppression.

Lane detection using spline model

Abstract In this paper, a Catmull–Rom spline-based lane model which describes the perspective effect of parallel lines has been proposed for generic lane boundary. Since Catmull–Rom spline can form arbitrary shapes by different sets of control points, it can describe a wider range of lane structures compared with other lane models, i.e. straight and parabolic models. Moreover, the lane detection problem has been formulated here as the problem of determining the set of control points of lane model. The proposed algorithm first detects the vanishing point (line) by using a Hough-like technique and then solves the lane detection problem by suggesting a maximum likelihood approach. Also, we have employed a multi-resolution strategy for rapidly achieving an accurate solution. This coarse-to-fine matching offers us an acceptable solution at an affordable computational cost, and thus speeds up the process of lane detection. As a result, the proposed method is robust to noise, shadows, and illumination variations in the captured road images, and is also applicable to both the marked and the unmarked roads.

Symmetry detection by generalized complex (GC) moments: a close-form solution

This paper presents a unified method for detecting both reflection-symmetry and rotation-symmetry of 2D images based on an identical set of features, i.e., the first three nonzero generalized complex (GC) moments. This method is theoretically guaranteed to detect all the axes of symmetries of every 2D image, if more nonzero GC moments are used in the feature set. Furthermore, we establish the relationship between reflectional symmetry and rotational symmetry in an image, which can be used to check the correctness of symmetry detection. This method has been demonstrated experimentally using more than 200 images.

Pattern recognition by graph matching using the Potts MFT neural networks

This paper is concerned with programming of the Potts mean field theory neural networks for pattern recognition by homomorphic mapping of the attributed relational graphs (ARG). In order to generate the homomorphic mapping from the scene relational graph to the model graph, we make use of the recently introduced [Suganthan, Technical Report, Nanyang Technical University (1994)] compatibility functions in relation to the Hopfield network. An efficient pose clustering algorithm is used to separate and localize different occurrences of any particular object model in the scene. The pose clustering algorithm also eliminates spurious hypotheses generated by the network and resolves ambiguities in the final interpretation. The performance of the proposed approach to pattern recognition by homomorphism is demonstrated using a number of line patterns, silhouette images and circle patterns.

Generalized 2D principal component analysis

A two-dimensional principal component analysis (2DPCA) by J. Yang et al. (2004) was proposed and the authors have demonstrated its superiority over the conventional principal component analysis (PCA) in face recognition. But the theoretical proof why 2DPCA is better than PCA has not been given until now. In this paper, the essence of 2DPCA is analyzed and a framework of generalized 2D principal component analysis (G2DPCA) is proposed to extend the original 2DPCA in two perspectives: a bilateral-projection-based 2DPCA (B2DPCA) and a kernel-based 2DPCA (K2DPCA) schemes are introduced. Experimental results in face recognition show its excellent performance.

A framework of 2D Fisher discriminant analysis: application to face recognition with small number of training samples

A novel framework called 2D Fisher discriminant analysis (2D-FDA) is proposed to deal with the small sample size (SSS) problem in conventional one-dimensional linear discriminant analysis (1D-LDA). Different from the 1D-LDA based approaches, 2D-FDA is based on 2D image matrices rather than column vectors so the image matrix does not need to be transformed into a long vector before feature extraction. The advantage arising in this way is that the SSS problem does not exist any more because the between-class and within-class scatter matrices constructed in 2D-FDA are both of full-rank. This framework contains unilateral and bilateral 2D-FDA. It is applied to face recognition where only few training images exist for each subject. Both the unilateral and bilateral 2D-FDA achieve excellent performance on two public databases: ORL database and Yale face database B.

Lane detection using B-snake

We propose a B-snake based lane detection algorithm. Compared with other lane models, the B-snake based lane model is able to describe a wider range of lane structures, since B-spline can form any arbitrary shape by a set of control points. The problems of detecting both sides of lane markings (or boundaries) have been formulated here as the problem of detecting the mid-line of the lane, by using the knowledge of the perspective parallel lines. A robust algorithm called CHEVP is presented for providing a good initial position for the B-snake. Furthermore, a minimum energy method by MMSE (minimum mean square energy) is suggested to determine the control points of the B-snake model by the overall image forces on two sides of lane. Experimental results show that the proposed method is robust against noise, shadows, and illumination variations in the captured road images, and also applicable to both the marked and the unmarked roads, and the dash and the solid paint line roads.

Pattern recognition by homomorphic graph matching using Hopfield neural networks

Abstract The application of the Hopfield neural network as a constraint satisfaction network for pattern recognition is investigated in this paper. Suitable energy and compatibility functions are introduced for pattern recognition by homomorphic attributed relational graph (ARG) matching. Although many computer vision problems have been traditionally formulated as combinatorial optimization problems, most of them can be reduced to that of finding the nearest local minimum of an objective function. In this paper, a novel network initialization strategy is applied to achieve the desired complexity reduction. Further, a method to verify and localize the hypotheses generated by the Hopfield network is also presented using an efficient pose clustering algorithm. The performance of the connectionist approach to pattern recognition by homomorphic relational graph matching is demonstrated using a number of line patterns, silhouette images and circle patterns.

A novel 3d partitioned active shape model for segmentation of brain MR images

A 3D Partitioned Active Shape Model (PASM) is proposed in this paper to address the problems of the 3D Active Shape Models (ASM). When training sets are small. It is usually the case in 3D segmentation, 3D ASMs tend to be restrictive. This is because the allowable region spanned by relatively few eigenvectors cannot capture the full range of shape variability. The 3D PASM overcomes this limitation by using a partitioned representation of the ASM. Given a Point Distribution Model (PDM), the mean mesh is partitioned into a group of small tiles. In order to constrain deformation of tiles, the statistical priors of tiles are estimated by applying Principal Component Analysis to each tile. To avoid the inconsistency of shapes between tiles, training samples are projected as curves in one hyperspace instead of point clouds in several hyperspaces. The deformed points are then fitted into the allowable region of the model by using a curve alignment scheme. The experiments on 3D human brain MRIs show that when the numbers of the training samples are limited, the 3D PASMs significantly improve the segmentation results as compared to 3D ASMs and 3D Hierarchical ASMs.

Bayesian shape model for facial feature extraction and recognition

Abstract A facial feature extraction algorithm using the Bayesian shape model (BSM) is proposed in this paper. A full-face model consisting of the contour points and the control points is designed to describe the face patch, using which the warping/normalization of the extracted face patch can be performed efficiently. First, the BSM is utilized to match and extract the contour points of a face. In BSM, the prototype of the face contour can be adjusted adaptively according to its prior distribution. Moreover, an affine invariant internal energy term is introduced to describe the local shape deformations between the prototype contour in the shape domain and the deformable contour in the image domain. Thus, both global and local shape deformations can be tolerated. Then, the control points are estimated from the matching result of the contour points based on the statistics of the full-face model. Finally, the face patch is extracted and normalized using the piece-wise affine triangle warping algorithm. Experimental results based on real facial feature extraction demonstrate that the proposed BSM facial feature extraction algorithm is more accurate and effective as compared to that of the active shape model (ASM).