| 基于Gabor特征的马铃薯图像芽眼识别 |
| |
| 引用本文： | 吕钊钦,亓协腾,张万枝,刘正铎,郑文秀,穆桂脂. 基于Gabor特征的马铃薯图像芽眼识别[J]. 农机化研究, 2021, 43(2): 203-207 |
| |
| 作者姓名： | 吕钊钦 亓协腾 张万枝 刘正铎 郑文秀 穆桂脂 |
| |
| 作者单位： | 山东农业大学机械与电子工程学院,山东泰安271018;山东省园艺机械与装备重点实验室,山东泰安271018;山东农业大学机械与电子工程学院,山东泰安271018;山东省园艺机械与装备重点实验室,山东泰安271018;山东农业大学机械与电子工程学院,山东泰安271018;山东省园艺机械与装备重点实验室,山东泰安271018;山东农业大学机械与电子工程学院,山东泰安271018;山东省园艺机械与装备重点实验室,山东泰安271018;山东农业大学机械与电子工程学院,山东泰安271018;山东省园艺机械与装备重点实验室,山东泰安271018;山东农业大学机械与电子工程学院,山东泰安271018;山东省园艺机械与装备重点实验室,山东泰安271018 |
| |
| 基金项目： | 山东省自然科学基金项目(ZR2019BC018);山东省农业重大应用技术创新项目(SD2019NJ003);山东省现代农业产业技术体系薯类创新团队农业机械岗位专家项目(SDAIT-16-10);“十三五”国家重点研发计划项目(2017YFD0700705);山东省农机装备研发创新计划项目(2018YF020-07,2017YF002)。 |
| |
| 摘 要： | 为了实现马铃薯智能化切块中芽眼自动识别,提出了一种基于Gabor特征的马铃薯图像芽眼识别方法。首先,设计了马铃薯图像采集系统,采集马铃薯图像;然后,进行马铃薯图像中芽眼特征分析,并基于Gabor特征进行马铃薯图像滤波,选择方向2和尺度4下的滤波图像进行形态学图像处理;最后,通过剔除马铃薯边界连通区域进行芽眼区域提取。选择50张马铃薯图像进行芽眼识别试验,结果表明:提出的算法芽眼识别率为93.4%,误识别率为7.2%,平均识别速度为0.28s。本算法可为马铃薯智能化切块中的芽眼自动识别提供方法。
|
| 关 键 词： | 马铃薯 芽眼识别 Gabor特征 图像处理 |
Buds Recognition of Potato Images Based on Gabor Feature |
| |
| Lv Zhaoqin,Qi Xieteng,Zhang Wanzhi,Liu Zhengduo,Zheng Wenxiu,Mu Guizhi. Buds Recognition of Potato Images Based on Gabor Feature[J]. Journal of Agricultural Mechanization Research, 2021, 43(2): 203-207 |
| |
| Authors: | Lv Zhaoqin Qi Xieteng Zhang Wanzhi Liu Zhengduo Zheng Wenxiu Mu Guizhi |
| |
| Affiliation: | (College of Mechanical and Electrical Engineering,Shandong Agricultural University,Tai’an 271018,China;Shandong Provincial Key Laboratory of Horticultural Machineries and Equipments,Tai’an 271018,China) |
| |
| Abstract: | In order to realize the automatic recognition of the buds in the intelligent dicing of potato,a method based on Gabor was proposed to identify the buds of potato images.Firstly,the potato image acquisition system is designed to collect the potato image;then the buds feature analysis in the potato image is carried out,and the potato image filtering is performed based on the Gabor feature,and the filtered image in the direction 2 and the scale 4 is selected for morphological image processing;finally,buds area can be extracted through excluding potato boundary connected areas.50 potato images were selected for the buds recognition test.The experimental results show that the proposed algorithm has a buds recognition rate of 93.4%,a false recognition rate of 14.2%,and an average recognition speed of 0.28 s.The algorithm in this paper can provide a method for automatic recognition of buds in potato intelligent dicing. |
| |
| Keywords: | potato buds recognition gabor feature image processing |
| 本文献已被 维普 万方数据 等数据库收录！ |
|
