The product configuration modeling is a key technology to realize configuration design. Most of the researches were mainly conducted focusing on how to establish a product configuration model, but were short of the co...The product configuration modeling is a key technology to realize configuration design. Most of the researches were mainly conducted focusing on how to establish a product configuration model, but were short of the consideration about the application scope and dynamic change features. In this paper, the reconfigurable modeling technology of similar-structure products configuration design is studied to enhance the generality and diversity of product configuration models, and to facilitate the dynamic upgrade of configuration models. The configuration components of products with similar structures are divided into sharable configuration components, essential configuration components and optional configuration components, so the construction of reconfigurable model of configuration design can be converted into the set operation of configuration components. The horizontal adjusting reconstruction of configuration model is carried out through the adding, replacing and deleting of configuration components, the vertical adjusting reconstruction of configuration model is carried out through the parametric deformation of configuration components, and the variable topological configuration model reconstruction can be realized through the transformation of partial structure of configuration components. Finally, NC machine products are used as an example to analyze and verify the reconfigurable modeling technology of products with similar structures. The speed of NC machine configuration design is found to be improved, and the number of NC machine configuration design models is decreased. And the reconfigurable design system of NC machine products is developed. The proposed modeling technology provides references for the processes, methods and steps of model reconstruction.展开更多
In recent years,addressing ill-posed problems by leveraging prior knowledge contained in databases on learning techniques has gained much attention.In this paper,we focus on complete three-dimensional(3D)point cloud r...In recent years,addressing ill-posed problems by leveraging prior knowledge contained in databases on learning techniques has gained much attention.In this paper,we focus on complete three-dimensional(3D)point cloud reconstruction based on a single red-green-blue(RGB)image,a task that cannot be approached using classical reconstruction techniques.For this purpose,we used an encoder-decoder framework to encode the RGB information in latent space,and to predict the 3D structure of the considered object from different viewpoints.The individual predictions are combined to yield a common representation that is used in a module combining camera pose estimation and rendering,thereby achieving differentiability with respect to imaging process and the camera pose,and optimization of the two-dimensional prediction error of novel viewpoints.Thus,our method allows end-to-end training and does not require supervision based on additional ground-truth(GT)mask annotations or ground-truth camera pose annotations.Our evaluation of synthetic and real-world data demonstrates the robustness of our approach to appearance changes and self-occlusions,through outperformance of current state-of-the-art methods in terms of accuracy,density,and model completeness.展开更多
This paper deals with the problem of clutter suppression in spaceborne distributed synthetic aperture radar （D-SAR） with nonuniform three-dimensional （3D） configuration geometry. In order to make a breakthrough of...This paper deals with the problem of clutter suppression in spaceborne distributed synthetic aperture radar （D-SAR） with nonuniform three-dimensional （3D） configuration geometry. In order to make a breakthrough of the configuration limitation of the traditional space time adaptive processing （STAP） based on uniform array and improve the inhomogeneous clutter suppres- sion performance, this paper considers signal reconstrtiction technique using array interpolation to process the D-SAR signal. An array interpolation signal reconstruction method based on pitching-partition is derived then a signal reconstruction 3D-STAP clutter suppression method applied to nonuniform 3D configuration is proposed. In particular, the proposed method is compared with conventional methods and the performance analysis is carried out based on simulations. The improvement factor （IF） for clutter suppression is imported and reported as a benchmark on the clutter suppression effect.展开更多
基金supported by National Natural Science Foundation of China (Grant No. 50905159)National S&T Major Project of China(Grant No. 2012ZX04010-011)Fundamental Research Funds for the Central Universities of China (Grant No. 2012FZA4001)
文摘The product configuration modeling is a key technology to realize configuration design. Most of the researches were mainly conducted focusing on how to establish a product configuration model, but were short of the consideration about the application scope and dynamic change features. In this paper, the reconfigurable modeling technology of similar-structure products configuration design is studied to enhance the generality and diversity of product configuration models, and to facilitate the dynamic upgrade of configuration models. The configuration components of products with similar structures are divided into sharable configuration components, essential configuration components and optional configuration components, so the construction of reconfigurable model of configuration design can be converted into the set operation of configuration components. The horizontal adjusting reconstruction of configuration model is carried out through the adding, replacing and deleting of configuration components, the vertical adjusting reconstruction of configuration model is carried out through the parametric deformation of configuration components, and the variable topological configuration model reconstruction can be realized through the transformation of partial structure of configuration components. Finally, NC machine products are used as an example to analyze and verify the reconfigurable modeling technology of products with similar structures. The speed of NC machine configuration design is found to be improved, and the number of NC machine configuration design models is decreased. And the reconfigurable design system of NC machine products is developed. The proposed modeling technology provides references for the processes, methods and steps of model reconstruction.
基金Supported by National Natural Science Foundation of China(Grant No.51935003).
文摘In recent years,addressing ill-posed problems by leveraging prior knowledge contained in databases on learning techniques has gained much attention.In this paper,we focus on complete three-dimensional(3D)point cloud reconstruction based on a single red-green-blue(RGB)image,a task that cannot be approached using classical reconstruction techniques.For this purpose,we used an encoder-decoder framework to encode the RGB information in latent space,and to predict the 3D structure of the considered object from different viewpoints.The individual predictions are combined to yield a common representation that is used in a module combining camera pose estimation and rendering,thereby achieving differentiability with respect to imaging process and the camera pose,and optimization of the two-dimensional prediction error of novel viewpoints.Thus,our method allows end-to-end training and does not require supervision based on additional ground-truth(GT)mask annotations or ground-truth camera pose annotations.Our evaluation of synthetic and real-world data demonstrates the robustness of our approach to appearance changes and self-occlusions,through outperformance of current state-of-the-art methods in terms of accuracy,density,and model completeness.
基金2011 China Aerospace Science and Technology Corporation Aerospace Science and Technology Innovation Foundation Subsidized Project2011 CASC and HIT Joint Technology Innovation Foundation
文摘This paper deals with the problem of clutter suppression in spaceborne distributed synthetic aperture radar （D-SAR） with nonuniform three-dimensional （3D） configuration geometry. In order to make a breakthrough of the configuration limitation of the traditional space time adaptive processing （STAP） based on uniform array and improve the inhomogeneous clutter suppres- sion performance, this paper considers signal reconstrtiction technique using array interpolation to process the D-SAR signal. An array interpolation signal reconstruction method based on pitching-partition is derived then a signal reconstruction 3D-STAP clutter suppression method applied to nonuniform 3D configuration is proposed. In particular, the proposed method is compared with conventional methods and the performance analysis is carried out based on simulations. The improvement factor （IF） for clutter suppression is imported and reported as a benchmark on the clutter suppression effect.
