Konvolutionäre neuronale Netze in der industriellen Bildverarbeitung und Robotik
In the first part of this dissertation, a framework for the design of a CNN for FPGAs is presented, consisting of a preprocessing algorithm, an augmentation technique, a custom quantization scheme and a pruning step of the CNN. The combination of conventional image processing with neural networks is...
Shranjeno v:
| Glavni avtor: | |
|---|---|
| Format: | Elektronski Book Chapter |
| Izdano: |
KIT Scientific Publishing
2022
|
| Serija: | Forschungsberichte aus der Industriellen Informationstechnik
26 |
| Teme: | |
| Online dostop: | OAPEN Library: download the publication OAPEN Library: description of the publication |
| Oznake: |
Označite
Brez oznak, prvi označite!
|
| Izvleček: | In the first part of this dissertation, a framework for the design of a CNN for FPGAs is presented, consisting of a preprocessing algorithm, an augmentation technique, a custom quantization scheme and a pruning step of the CNN. The combination of conventional image processing with neural networks is shown in the second part by an example from robotics, where an image-based visual servoing process is successfully conducted for a gripping process of a robot. |
|---|---|
| Fizični opis: | 1 electronic resource (212 p.) |
| ISBN: | KSP/1000146397 |
| Dostop: | Open Access |