A study of the sorting device for Chinese cabbage seeds with machine vision
| Autor: | Jian-Feng Cheng, 程建峰 |
|---|---|
| Rok vydání: | 2015 |
| Druh dokumentu: | 學位論文 ; thesis |
| Popis: | 103 In this study, an auto-sorting system for detecting and sorting Chinese cabbage seeds is developed because Chinese cabbage seeds are sorted manually. The functions of device include auto-feeding, image capture, classification and seeds collection. In this study, an application of neural network and image processing techniques were used to detect and classify Chinese cabbage seeds. The size, shape, color and texture features of seeds were estimated using image processing. The seeds classifier was proposed to classify seeds according to the features of seeds. The grade of seed includes excellent, good and bad. The accuracies of classification were 91.53 % and 88.95 % for excellent and good seeds. The average accuracy was 90.38 %. The speed of sorting arrived to 200 seeds in a minute. The testing results show that the Chinese cabbage seeds can be sorted using the developed system efficiently. |
| Databáze: | Networked Digital Library of Theses & Dissertations |
| Externí odkaz: |
|