| Autor: |
Ai Xia Guo, Xiangjun Zou, Jun Chao Xiong, Li Juan Chen, Hong Xing Peng, Ding Zhong Wu |
| Rok vydání: |
2012 |
| Předmět: |
|
| Zdroj: |
Key Engineering Materials. 522:720-725 |
| ISSN: |
1662-9795 |
| DOI: |
10.4028/www.scientific.net/kem.522.720 |
| Popis: |
The problem of robot operation based on vision is to locate and identify the target. The approach of active shape model has been used to identify the target and taken citrus as example in this article. First, the different shapes of citrus are calibrated and calibrated; Second, the method of principal component analysis is used to get the models of different shapes of citrus; Finally, the models are used to do gray matching with the actual citrus and the characteristic parameters such as the ratio of pixel to number are extracted to achieve the identify of the citrus image. The experimental results show that the method has higher recognition rate for citrus image and better robust and good flexibility. |
| Databáze: |
OpenAIRE |
| Externí odkaz: |
|
