Showing papers by "Rushi Lan published in 2011"
25 Apr 2011
TL;DR: In this paper, a cross-weighted centroid (CWC) is constructed to extract affine invariant features, which is derived from these weighted points, and a series of centroids of some affine regions can be obtained by iterating affine region cutting, i.e., ARC.
Abstract: In this paper, cross-weighted centroid (CWC) is constructed to extract affine invariant features. Every point in an image is assigned with a cross-weight based on the distribution of the image. CWC, derived from these weighted points, is affine invariance. Based on the original centroid and CWC of the image, a series of centroids of some affine regions can be obtained by iterating affine region cutting, i.e., ARC. Several affine invariant triangles are constructed. Consequently, affine invariant features can be derived by the area of these triangles. Experiment results show the efficiency of the proposed method.
1 citations
26 Mar 2011
TL;DR: This paper describes a corner detection method in a pseudo-random structured light pattern in which the symmetry property between two neighboring rhombic elements is considered and the fuzzy c-means (FCM) algorithm is conducted to determine the threshold for being the corner.
Abstract: This paper describes a corner detection method in a pseudo-random structured light pattern. In the algorithm, the image is firstly convoluted with a weighted Gaussian mask (WGM) in which the symmetry property between two neighboring rhombic elements is considered. As a result, the proposed method is more suitable to detect the X shape corner in the structured light pattern. Then a non-maximal suppression process is carried out to find the candidates for the corner. Record the times of each position being the candidate in different size of WGMs. Finally, the fuzzy c-means (FCM) algorithm is conducted to determine the threshold for being the corner. Some experiments have been conducted to demonstrate the effectiveness of the proposed method.
25 Apr 2011
TL;DR: In the proposed method, weighted cross mask convolution and non-maximal suppression are carried out to the pattern to obtain several grid-point candidates and Harris corner detector is used to accurately measure the point symmetry.
Abstract: In this paper, an adaptive grid-point detection method is developed in a pseudo-random color pattern. In the proposed method, weighted cross mask convolution and non-maximal suppression are carried out to the pattern to obtain several grid-point candidates. In order to accurately measure the point symmetry of the candidates, Harris corner detector is used to adaptively select the suitable window size. Finally, the grid-points are recognized by comparing the symmetry characteristic with a given threshold. Some experiments have been conducted to demonstrate the effectiveness of the proposed method.