| 基于迁移学习的GoogLenet煤矸石图像识别 |
| |
| 引用本文： | 曹现刚,薛祯也.基于迁移学习的GoogLenet煤矸石图像识别[J].教育技术导刊,2019,18(12):183-186. |
| |
| 作者姓名： | 曹现刚 薛祯也 |
| |
| 作者单位： | 西安科技大学 机械工程学院，陕西 西安 710054 |
| |
| 基金项目： | 陕西省重点研发计划项目（2018GY-160） |
| |
| 摘 要： | 已有的煤矸石识别方法具有一定效果，但无法满足实际需求。为了寻找新的煤矸石识别方法，提出了基于深度学习的煤矸石图像识别方法。采用Inception模型，并通过迁移学习共享已训练模型卷积层权值和偏差。从煤矸石图像库中随机抽取煤矸石图像作为训练集和测试集，最后将该方法与传统图像识别方法进行比对。实验结果表明，该模型可以有效识别煤矸石，准确率为93.5%，有效提高了煤矸识石别准确率。
|
| 关 键 词： | GoogLenet 煤矸石识别 迁移学习 |
| 收稿时间： | 2019-03-08 |
Coal Gangue Identification by Using Transfer Learning in GoogLenet |
| |
| CAO Xian-gang,XUE Zhen-ye.Coal Gangue Identification by Using Transfer Learning in GoogLenet[J].Introduction of Educational Technology,2019,18(12):183-186. |
| |
| Authors: | CAO Xian-gang XUE Zhen-ye |
| |
| Institution: | College of Mechanical Engineering，Xi’an University of Science and Technology，Xi’an 710054，China |
| |
| Abstract: | The existing coal gangue identification method has achieved certain results， but it cannot meet the actual demand. In order to find a new coal shovel identification method， a coal gangue image recognition method based on deep learning is proposed. By using the Inception model， the shared model convolution layer weights and deviations are shared by learning through migration. The coal gangue image is randomly selected from the coal gangue image database as the training set and test set. Finally， the method is compared with the traditional image recognition method. The results show that the model can effectively identify coal gangue with an accuracy of 93.5%， which effectively improves the accuracy of coal gangue identification. |
| |
| Keywords: | coal gangue Identification GoogLenet transfer learning |
|
| 点击此处可从《教育技术导刊》浏览原始摘要信息 |
| 点击此处可从《教育技术导刊》下载免费的PDF全文 |
