Optimising connectionist models and attributed relational graph matching for object recognition
This research work describes in depth investigation into optimising connectionist models and their applications in rigid object and pattern recognition by attributed relational graph (ARG) matching. The ARG representation is chosen because it encodes relational semantic information in itself and per...
Saved in:
| 主要作者: | |
|---|---|
| 其他作者: | |
| 格式: | Theses and Dissertations |
| 語言: | English |
| 出版: |
2009
|
| 主題: | |
| 在線閱讀: | http://hdl.handle.net/10356/19661 |
| 標簽: |
添加標簽
沒有標簽, 成為第一個標記此記錄!
|
| 機構: | Nanyang Technological University |
| 語言: | English |
| 總結: | This research work describes in depth investigation into optimising connectionist models and their applications in rigid object and pattern recognition by attributed relational graph (ARG) matching. The ARG representation is chosen because it encodes relational semantic information in itself and performs well under clutter and partial occlusion. The matching of model and scene ARGs is performed using optimising con-nectionist models. Since the connectionist models offer parallel and distributed process-ing, and cost effective hardware implementation, optimising connectionist model-based recognition systems can be employed to solve practical recognition problems. |
|---|