OALib Journal期刊

ISSN: 2333-9721




2018 ( 2 )

2017 ( 132 )

2016 ( 161 )

2015 ( 1935 )


匹配条件: “Wu Fu-chao” ,找到相关结果约64202条。
Studies on Cover Algorithm of Classification Decision

Yang Jin-fu,Wu Fu-chao,

电子与信息学报 , 2007,
Abstract: Data-based machine learning is exploring the rule to predict new data from the observation data. In this paper, a novel classification decision method, called as Cover Algorithm (CA), is presented. In the training procedure, some representative samples of the training set can be obtained by utilizing a certain cover rule. Then, in the classification phase, the classifier can make a decision according to the distances from a test sample to the representatives, namely the class of the test sample is determined by the representative closest to the test sample. Comparing with the nearest neighbor method, the presented method needs less cost and memory space as the representative samples are only a little part of the training set. Furthermore, cover algorithm is suitable for automated classification of large data because it does not need to consider choosing kernel function like SVM and its main computation is distance operation between samples. The experiment results show that the cover algorithm has good robustness and high classifying accuracy over Normal Galaxies and stars datasets.

WU Fu-Chao,Hu Zhan-Yi,

自动化学报 , 2002,
Abstract: Homography estimation is widely used for 3D motion analysis, mosaicing, camera calibration and more. This paper is concentrated on the investigation on multi view and multi plane constraints on homographies. The main results are: A. when the camera is under a pure translation, then, 1) both the homographies of all planes between two view points, and the homographies of a plane for all view points are of rank 4, 2) the standard homographies in multi view and multi plane case are still of rank 4, 3) the above two results keep true for the relative homographies in the case of general camera motion; B. when the camera is under a general motion, 1) the homographies of all planes between two view points are also of rank 4, 2) however, the rank should be 9 rather than 4 in other cases.
Multi-Camera Calibration Based on 1D Calibration Object

WANG Liang,WU Fu-Chao,

自动化学报 , 2007,
Abstract: A one-dimensional calibration object consists of three or more collinear points with known relative positions. It is generally believed that a camera can be calibrated only when a 1D calibration object is in planar motion or rotates around a fixed point. In this paper, it is proved that when a multi-camera is observing a 1D object undergoing general rigid motions synchronously, the camera set can be linearly calibrated. A linear algorithm for the camera set calibration is proposed, and then the linear estimation is further refined using the maximum likelihood criteria. The simulated and real image experiments show that the proposed algorithm is valid and robust.
5-point and 4-point Algorithm to Determine of the Fundamental Matrix

WU Fu-Chao,HU Zhan-Yi,

自动化学报 , 2003,
Abstract: The fundamental matrix encapsulates all the information between two images, and plays a very important role in camera calibration and 3D reconstruction. In this paper, the following conclusions have been rigorously proved: If the camera motion is of a pure translation, then given 5 point correspondences across two images, the fundamental matrix can be linearly determined if four correspondences of the 5 ones are from coplanar space points (called coplanar correspondences). In addition, we show that if the distortion factor in the pinhole camera model is null, then the fundamental matrix can be linearly determined by only these 4 coplanar correspondences. To our knowledge, such results have not been reported yet in the literature.

WU Fu-Chao,HU Zhan-Yi,

自动化学报 , 2002,
Abstract: In this paper, a new constraint on the homography of the plane at infinity is introduced and a new linear camera calibration technique is proposed based on it. Compared with the related techniques in the literature, the main advantages of our new technique are tow-fold. Firstly, it is less stringent to hardware, for example, it does not require the camera to undertake orthogonal motions which are usually difficult to be done without special hardware support. In contrast, our technique requires only one translation, and two general motions of camera, which can be easily done, for example, with a hand-held camera. Secondly, in the determination of the homography of the plane at infinity, it relies neither on projective reconstruction nor on the homography of a space plane, it needs only some image correspondences and epipoles, which are basic requirements for any camera self-calibration technique. In addition, we prove that for a given set of camera motions such as {(R,t1),(R,t2)},if (t1,t2) are not linearly dependent, then the homography of the plane at infinity under this motion set can be linearly and uniquely determined. Simulations and experiments with real images validate our new method.
Metric Reconstruction Based on One Triplet of Corresponding Vanishing Line

ZHU Hai-Jiang,WU Fu-Chao,

软件学报 , 2004,
Abstract: A technique for metric reconstruction based on one triplet of corresponding vanishing line in three images is proposed in this paper. Firstly, the infinite homography is computed by using the vanishing line correspondence and the modulus constraint. Then the camera intrinsic parameters are determined linearly by using the property that the infinite homography preserves the image of the absolute conic. Finally a metric reconstruction is obtained. Experiments with simulated data as well as with real images show that the proposed method is workable and applicable in real applications.
Pseudosphere Filter and Edge Detection

WANG Zhi-Heng,WU Fu-Chao,

软件学报 , 2008,
Abstract: An image filter called the pseudosphere filter is presented in this paper.An edge-preserving parameter is introduced besides a scale parameter in the pseudosphere filter,and thus a better trade-off between image smoothing and edge locating can be obtained by using it.A pseudosphere-based edge detector is built by replacing the Gaussian filter in the classic Canny edge detector with the pseudosphere filter.Experimental results show that, by comparing with the classic Canny edge detector,in the case of having the same smoothness,pseudosphere-based edge detector offers a better precision for edge locating.
A Linear Method for the PnP Problem

WU Fu-Chao,HU Zhan-Yi,

软件学报 , 2003,
Abstract: 经典的PnP(3(n(5)问题从本质上来说是非线性的,不但具有多解性而且对图像点的位置误差极为敏感.经典PnP问题仅涉及一幅图像,针对机器人导航中的实际需求,将PnP问题扩展到了摄像机作平移运动下的两幅图像的情况,并研究探讨了在这种情况下PnP问题的线性求解方法,主要结果有:给定平移运动下n个控制点的两幅图像,(1) 当n=3时,可线性求解摄像机的方位以及摄像机内参数的两个尺度因子;(2) 当n4时,不仅可以线性求解摄像机的方位,而且能够确定摄像机的所有内参数.也就是说,给定平移视点下的两幅控制点图像,可以线性求解摄像机未标定的PnP问题.结果具有一定的理论意义和应用价值.
The Number of Independent Kruppa Constraints from N Images
Zhan-Yi Hu,Yi-Hong Wu,Fu-Chao Wu,Song-De Ma,
Zhan-Yi Hu
,Yi-Hong Wu,Fu-Chao Wu,and Song-De Ma

计算机科学技术学报 , 2006,
Abstract: It is well known that without any priori knowledge on the scene, camera motion and camera intrinsic parameters, the only constraint between a pair of images is the so-called epipolar constraint, or equivalently its fundamental matrix. For each fundamental matrix, at most two independent constraints on the cameras' intrinsic parameters are available via the Kruppa equations. Given N images, N(N - 1)/2 fundamental matrices appear, and N(N - 1) Kruppa constraints are available. However, to our knowledge, a formal proof of how many independent Kruppa constraints exist out of these N(N - 1) ones is unavailable in the literature. In this paper, we prove that given N images captured by a pinhole camera with varying parameters and under general motion, the number of independent Kruppa constraints is (5N - 9) (N > 2), and it is less than that of independent constraints from the absolute quadric by only one. This one-constraint-less property of the Kruppa equations is their inherent deficiency and is independent of camera motion. This deficiency is due to their failure of automatic enforcement of the rank-three-ness on the absolute quadric.
New Active Vision Based Camera Self-Calibration Technique

WU Fu-Chao,LI Hua,HU Zhan-Yi,

自动化学报 , 2001,
Abstract: In this paper, a new active vision based camera self calibration technique is proposed. The novelty of this new technique is that it can determine LINEARLY all the FIVE intrinsic parameters of the camera. The basic principle of our new calibration technique is to use the planar information in the scene and to control the camera to undergo several sets of orthogonal planar motions. Then, a set of linear constraints on the five intrinsic parameters is derived by means of homographies associated with scene planes.In addition,the uniqueness of the solution of the constraint set with respect to the configurations of the camera's planar motions is also investigated.

Copyright © 2008-2017 Open Access Library. All rights reserved.