Cherry recognition in natural environment based on the vision of picking robot
| Autor: | Tingzhong Yu, Qirong Zhang, Yan Wang, Shanxiong Chen |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
Computer science
business.industry 0211 other engineering and technologies Process (computing) Centroid 02 engineering and technology Hough transform law.invention Image (mathematics) 020303 mechanical engineering & transports 0203 mechanical engineering law 021105 building & construction Median filter Robot Segmentation Computer vision Noise (video) Artificial intelligence business |
| Zdroj: | IOP Conference Series: Earth and Environmental Science. 61:012021 |
| ISSN: | 1755-1315 1755-1307 |
| DOI: | 10.1088/1755-1315/61/1/012021 |
| Popis: | In order to realize the automatic recognition of cherry in the natural environment, this paper designed a robot vision system recognition method. The first step of this method is to pre-process the cherry image by median filtering. The second step is to identify the colour of the cherry through the 0.9R-G colour difference formula, and then use the Otsu algorithm for threshold segmentation. The third step is to remove noise by using the area threshold. The fourth step is to remove the holes in the cherry image by morphological closed and open operation. The fifth step is to obtain the centroid and contour of cherry by using the smallest external rectangular and the Hough transform. Through this recognition process, we can successfully identify 96% of the cherry without blocking and adhesion. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|