RGB-D-Based Features for Recognition of Textureless Objects
| Autor: | Santosh Thoduka, Stepan Pazekha, Gerhard K. Kraetzschmar, Alexander Moriarty |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
business.industry
Computer science 020209 energy ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Point cloud Cognitive neuroscience of visual object recognition 02 engineering and technology Set (abstract data type) Range (mathematics) 020303 mechanical engineering & transports 0203 mechanical engineering Minimum bounding box 0202 electrical engineering electronic engineering information engineering RGB color model Robot Computer vision Artificial intelligence business Resolution (algebra) |
| Zdroj: | RoboCup 2016: Robot World Cup XX ISBN: 9783319687919 RoboCup |
| DOI: | 10.1007/978-3-319-68792-6_24 |
| Popis: | Autonomous industrial robots need to recognize objects robustly in cluttered environments. The use of RGB-D cameras has progressed research in 3D object recognition, but it is still a challenge for textureless objects. We propose a set of features, including the bounding box, mean circle fit and radial density distribution, that describe the size, shape and colour of objects. The features are extracted from point clouds of a set of objects and used to train an SVM classifier. Various combinations of the proposed features are tested to determine their influence on the recognition rate. Medium-sized objects are recognized with high accuracy whereas small objects have a lower recognition rate. The minimum range and resolution of the cameras are still an issue but are expected to improve as the technology improves. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|