Knygos.lt nariams
Knygos.lt nariams
- Išsiųsime per 12–18 d.d.
/images/books/13376082/9783838131405.jpg)
- Autorius: Nikolai Sergeev
- Leidėjas: Sudwestdeutscher Verlag Fur Hochschulschriften AG
- Metai: 2012
- Puslapiai: 104
- ISBN-10: 3838131401
- ISBN-13: 9783838131405
- Formatas: 15.2 x 22.9 x 0.6 cm, minkšti viršeliai
- Kalba: Anglų
Atsiliepimai
Aprašymas
In this book a new approach to object recognition will be introduced. The basic idea of the approach is to compare two combinations of half ellipses. It is to find out whether the combinations can be at least partially transformed in each other through an affine mapping. As the transformation can be partial the system is robust to occlusion. The transformation does not have to be exact. Two combinations being able to be transformed in each other approximately with an error up to some epsilon-bound are regarded as similar. The epsilon-error tolerance makes the system robust to deformation. Further attractive properties of the approach are: capability to stable separation of a single object from its background or of several objects partially occluding each other from one another, capability to learn a new object in a time independent of the number of objects already learned, color information can be ignored or combined with form representation. Additionally a new near neighbor search algorithm developed to handle the representation based on half ellipses will be introduced.
- Autorius: Nikolai Sergeev
- Leidėjas: Sudwestdeutscher Verlag Fur Hochschulschriften AG
- Metai: 2012
- Puslapiai: 104
- ISBN-10: 3838131401
- ISBN-13: 9783838131405
- Formatas: 15.2 x 22.9 x 0.6 cm, minkšti viršeliai
- Kalba: Anglų
In this book a new approach to object recognition will be introduced. The basic idea of the approach is to compare two combinations of half ellipses. It is to find out whether the combinations can be at least partially transformed in each other through an affine mapping. As the transformation can be partial the system is robust to occlusion. The transformation does not have to be exact. Two combinations being able to be transformed in each other approximately with an error up to some epsilon-bound are regarded as similar. The epsilon-error tolerance makes the system robust to deformation. Further attractive properties of the approach are: capability to stable separation of a single object from its background or of several objects partially occluding each other from one another, capability to learn a new object in a time independent of the number of objects already learned, color information can be ignored or combined with form representation. Additionally a new near neighbor search algorithm developed to handle the representation based on half ellipses will be introduced.
Atsiliepimai