This paper introduces one passive prism based single-lens multi-ocular stereo image capture system. It only employs one real CCD camera with a pyramid-like multi-face glass prism. Each image captured by this system ca...
详细信息
This paper introduces one passive prism based single-lens multi-ocular stereo image capture system. It only employs one real CCD camera with a pyramid-like multi-face glass prism. Each image captured by this system can be divided into three, four or more sub-images which can be taken as the images simultaneously captured by a group of virtual cameras generated by the prism. Hence this system can be further categorized into trinocular, quadric-ocular, penta-ocular and other multi-ocular sub-systems depending on the number of virtual cameras. Two different approaches are developed to understand this system such that the stereo information embedded in each image can be recovered: one approach is based on camera calibration technique and another is based on geometrical analysis on ray sketching. Experiments are conducted to test the feasibilities of the both approaches
作者：
Oliver NellesAutomatic Control
Mechatronics Department of Mechanical Engineering University of Siegen Siegen Germany
Local model networks, also known as Takagi-Sugeno neuro-fuzzy systems, have become an increasingly popular nonlinear model architecture. Usually the local models are linearly parameterized and those parameters are typ...
详细信息
Local model networks, also known as Takagi-Sugeno neuro-fuzzy systems, have become an increasingly popular nonlinear model architecture. Usually the local models are linearly parameterized and those parameters are typically estimated by some least squares approach. However, widely different strategies have been pursued for the partitioning of the input space which determines the validity regions of the local models. The model properties crucially depend on the chosen strategy. This paper proposes an axes-oblique partitioning strategy and an efficient construction algorithm for its realization. Many advantages over the existing approaches are demonstrated
In this paper, we introduce the Bayesian kernel inference to the classical problem of 2D objects recognition using the normalized curvature as the feature vector. The idea and formulations of curvature normalization a...
详细信息
In this paper, we introduce the Bayesian kernel inference to the classical problem of 2D objects recognition using the normalized curvature as the feature vector. The idea and formulations of curvature normalization are proposed. A Gaussian window is applied to the feature before the normalization. The proposed Bayesian classifier in the hyperspace is formed by dramatically few kernels, which is significant for the problem with big learning sample database. The experiments show that this method has an excellent accuracy and insensitive to the noise. This algorithm also could easily be applied to multi-class problems
In this paper, we present an innovative compact discrete polar coordinates transform for the identification and restoration of rotational blurred image. A relaxed solution of the transform function and its inverse is ...
详细信息
In this paper, we present an innovative compact discrete polar coordinates transform for the identification and restoration of rotational blurred image. A relaxed solution of the transform function and its inverse is presented in discrete domain. Based on analysis and assumptions, interpolation is applied in the discrete polar domain to solve the problem of under sampling. The cost function of the compactness is introduced. Based on the experimental results, this coordinates transform can simplify the spatial variant problem to a spatial invariant problem
In this paper, we examine a linear one degree of freedom engine mount to obtain optimum mount parameters in a passive configuration. An engine mount is a device that may be used to isolate vehicle body from the engine...
详细信息
A novel lapping mode is developed on the base of present eccentric v-grooves lapping mode for precise balls. The kinematics analysis shows that the ball spin axis can wiggle up to 180° under the developed mode, s...
详细信息
A novel lapping mode is developed on the base of present eccentric v-grooves lapping mode for precise balls. The kinematics analysis shows that the ball spin axis can wiggle up to 180° under the developed mode, so the lapping trace will be dispersed onto the whole surface of the ball. The effect of offsetting and groove radius on the lapping uniformity in the developed eccentric v-grooves, the present eccentric v-grooves and the traditional v-grooves lapping mode are compared, and the advantage of the developed mode is that the impact offsetting is less while the groove radius has no impact.
This paper introduces one passive prism based single-lens multi-ocular stereo image capture *** only employs one real CCD camera with a pyramidlike multi-face glass *** image captured by this system can be divided int...
详细信息
This paper introduces one passive prism based single-lens multi-ocular stereo image capture *** only employs one real CCD camera with a pyramidlike multi-face glass *** image captured by this system can be divided into three,four or more subimages which can be taken as the images simultaneously captured by a group of virtual cameras generated by the *** this system can be further categorized into trinocular,quadric-ocular,penta-ocular and other multi-ocular sub-systems depending on the number of virtual *** different approaches are developed to understand this system such that the stereo information embedded in each image can be recovered:one approach is based on camera calibration technique and another is based on geometrical analysis on ray *** are conducted to test the feasibilities of the both approaches.
In this paper,we will present an innovative compact discrete polar coordinates transform for the identification and restoration of rotational blurred image.A relaxed solution of the transform function and its inverse ...
详细信息
In this paper,we will present an innovative compact discrete polar coordinates transform for the identification and restoration of rotational blurred image.A relaxed solution of the transform function and its inverse is presented in discrete *** on analysis and assumptions,interpolation is applied in the discrete polar domain to solve the problem of *** cost function of the compactness is *** on the experimental results,this coordinates transform can simplify the spatial variant problem to a spatial invariant problem.
In this paper,we introduce the Bayesian kernel inference to the classical problem of 2D objects recognition using the normalized curvature as the feature *** idea and formulations of curvature normalization are propos...
详细信息
In this paper,we introduce the Bayesian kernel inference to the classical problem of 2D objects recognition using the normalized curvature as the feature *** idea and formulations of curvature normalization are proposed.A Gaussian window is applied to the feature before the *** proposed Bayesian classifier in the hyperspace is formed by dramatically few kernels,which is significant for the problem with big learning sample *** experiments show that this method has an excellent accuracy and insensitive to the *** algorithm also could easily be applied to multi-class problems.
暂无评论