Accurate 3D geometry measurement for non-cooperative spacecraft with an unfocused light-field camera

This work explores an alternative 3D geometry measurement method for non-cooperative spacecraft guiding navigation and proximity operations.From one snapshot of an unfocused light-field camera, the 3D point cloud of a non-cooperative spacecraft can be calculated from sub-aperture images with the epipolar plane image(EPI) based light-field rendering algorithm.A Chang'e-3 model(7.2 cm×5.6 cm×7.0 cm) is tested to validate the proposed technique.Three measurement distances(1.0 m, 1.2 m, 1.5 m) are considered to simulate different approaching stages.Measuring errors are quantified by comparing the light-field camera data with a high precision commercial laser scanner.The mean error distance for the three cases are 0.837 mm, 0.743 mm, and 0.973 mm respectively, indicating that the method can well reconstruct 3D geometry of a non-cooperative spacecraft with a densely distributed 3D point cloud and is thus promising in space-related missions.

Urban scene recognition by graphical model and 3D geometry

This paper proposes a simple and discriminative framework, using graphical model and 3D geometry to understand the diversity of urban scenes with varying viewpoints. Our algorithm constructs a conditional random field （CRF） network using over-segmented superpixels and learns the appearance model from different set of features for specific classes of our interest. Also, we introduce a training algorithm to learn a model for edge potential among these superpixel areas based on their feature difference. The proposed algorithm gives competitive and visually pleasing results for urban scene segmentation. We show the inference from our trained network improves the class labeling performance compared to the result when using the appearance model solely.

3D Modelling of Biological Systems for Biomimetics

With the advanced development of computer-based enabling technologies, many engineering, medical, biology, chemistry, physics and food science etc have developed to the unprecedented levels, which lead to many research and development interests in various multi-discipline areas. Among them, biomimetics is one of the most promising and attractive branches of study. Biomimetics is a branch of study that uses biological systems as a model to develop synthetic systems. To learn from nature, one of the fundamental issues is to understand the natural systems such animals, insects, plants and human beings etc. The geometrical characterization and representation of natural systems is an important fundamental work for biomimetics research. 3D modeling plays a key role in the geometrical characterization and representation, especially in computer graphical visualization. This paper firstly presents the typical procedure of 3D modelling methods and then reviews the previous work of 3D geometrical modelling techniques and systems developed for industrial, medical and animation applications. Especially the paper discusses the problems associated with the existing techniques and systems when they are applied to 3D modelling of biological systems. Based upon the discussions, the paper proposes some areas of research interests in 3D modelling of biological systems and for Biomimetics.

A Skeletal Camera Network for Close-range Images with a Data Driven Approach in Analyzing Stereo Configuration

Structure-from-Motion(SfM)techniques have been widely used for 3D geometry reconstruction from multi-view images.Nevertheless,the efficiency and quality of the reconstructed geometry depends on multiple factors,i.e.,the base-height ratio,intersection angle,overlap,and ground control points,etc.,which are rarely quantified in real-world applications.To answer this question,in this paper,we take a data-driven approach by analyzing hundreds of terrestrial stereo image configurations through a typical SfM algorithm.Two main meta-parameters with respect to base-height ratio and intersection angle are analyzed.Following the results,we propose a Skeletal Camera Network(SCN)and embed it into the SfM to lead to a novel SfM scheme called SCN-SfM,which limits tie-point matching to the remaining connected image pairs in SCN.The proposed method was applied in three terrestrial datasets.Experimental results have demonstrated the effectiveness of the proposed SCN-SfM to achieve 3D geometry with higher accuracy and fast time efficiency compared to the typical SfM method,whereas the completeness of the geometry is comparable.

Finocyl Grain Design and Optimization Using Sequential Quadratic Programming

Design technique of 3D Finocyl grain configuration for Solid Rocket Motors, including its performance prediction and optimization is discussed. In doing so, the design objectives and constraints are set, geometric parameters of Finocyl grain are identified, and performance prediction parameters are calculated, thereafter the preliminary design is completed and optimal design is reached. For every grain design, it is necessary that the minimum possible mass of propellant is used to produce the required thrust within a certain limit of burning time. By using this technique of design and optimization, the vital parameter of propellant mass is optimized to its minimum value, yet vital parameter of thrust is attained in the required burning time with the fixed length and diameter of motor. Especially a geometrical model of grain configuration is developed by using various combinations of ellipsoid, cone, cylinder, sphere, torus and inclined plane. With the diameter of the motor fixed, the Finocyl Grain geometry totally depends on sixteen independent variables. Each of these variables has a bearing on explicit characteristic of Finocyl grain design and optimization. Changing the value of each of these variables brings significant effects on the performance. Due to such attributes of Finocyl grain configuration, compromises will result. Overall optimal design is ensured through assigning and analyzing a suitable range of geometric parameters satisfying the requirements of minimum mass of propellant and ensuring sound values for internal ballistic parameters while remaining within the design constraints of thrust, burning time, length and diameter of chamber case.

Swirling flows in horizontally vibrated beds of dense granular materials

In a series of experiments, a granular material in a rectangular container with two hollow cylinders was studied as it underwent horizontal vibrations. At the peak values of acceleration, novel swirling granular flows were observed in the cylinders while the grains cascaded down the outer surface of the piles that formed outside the cylinders. Computer simulations were performed that supported our interpretation of the behaviour observed in the experiments.