IEEE Transactions on Systems Man and Cybernetics Part B (Cybernetics) · 2004 · 185 citations · 36 references
EngineeringComputer ScienceRobotics EngineeringComputer VisionMachine VisionRobot Vision3D ReconstructionGeometry ProcessingMappingMulti-view GeometryRobotic VisionScene ReconstructionGenetic Algorithm OptimizationObject ModelingPath ComputationSensor PlacementSensor Placement DesignSensor Placement OptimizationShortest Path OptimizationShortest Path Planning
In model‑based robot vision, the sensor must be repositioned around an object to capture all relevant features, requiring multiple 3‑D images from diverse viewpoints. This work proposes an automatic sensor‑placement method that identifies optimal viewpoints and computes a minimal‑path route for the sensor. The planner resamples object surface points with normals, builds an optimal‑placement graph optimized by a genetic algorithm using a min‑max criterion, and finds a shortest path with Christofides’ algorithm, all implemented in a Viewpoint Planner that supports animation, specification, and simulation. Experiments on a real robot vision system confirm that the proposed method effectively generates efficient sensor‑placement plans.
This paper presents a method for automatic sensor placement for model-based robot vision. In such a vision system, the sensor often needs to be moved from one pose to another around the object to observe all features of interest. This allows multiple three-dimensional (3-D) images to be taken from different vantage viewpoints. The task involves determination of the optimal sensor placements and a shortest path through these viewpoints. During the sensor planning, object features are resampled as individual points attached with surface normals. The optimal sensor placement graph is achieved by a genetic algorithm in which a min-max criterion is used for the evaluation. A shortest path is determined by Christofides algorithm. A Viewpoint Planner is developed to generate the sensor placement plan. It includes many functions, such as 3-D animation of the object geometry, sensor specification, initialization of the viewpoint number and their distribution, viewpoint evolution, shortest path computation, scene simulation of a specific viewpoint, parameter amendment. Experiments are also carried out on a real robot vision system to demonstrate the effectiveness of the proposed method.
36
Viewpoint Selection using Viewpoint Entropy
Pere‐Pau Vázquez, Miquel Feixas, Mateu Sbert et al. · 2001 · 317 citations