| 基于轮廓特征的单只蛋鸡行为识别方法 |
| |
| 引用本文： | 赵守耀,陆辉山,王福杰,李沛,王宁.基于轮廓特征的单只蛋鸡行为识别方法[J].中国农机化学报,2022(2):143-147,181. |
| |
| 作者姓名： | 赵守耀 陆辉山 王福杰 李沛 王宁 |
| |
| 作者单位： | 中北大学机械工程学院 |
| |
| 基金项目： | 国家重点研发计划项目(2016YFD0700202)。 |
| |
| 摘 要： | 育成期蛋鸡的行为能够反映其健康状况和对环境的适应情况,为快速地识别蛋鸡的行为,提出一种基于蛋鸡轮廓特征的行为识别方法.首先获取蛋鸡俯视图图像,并对蛋鸡图像进行预处理,通过自动阈值分割法得到蛋鸡轮廓并进行拟合,提取出拟合轮廓的几何特征,然后对特征进行排列组合,得到四种特征组合并结合极限学习机(ELM)进行训练,得到最佳的...
|
| 关 键 词： | 行为识别 特征提取 图像处理 极限学习机 |
Recognition method of single layer behavior based on contour feature |
| |
| Zhao Shouyao,Lu Huishan,Wang Fujie,LI Pei,Wang Ning.Recognition method of single layer behavior based on contour feature[J].Chinese Agricultural Mechanization,2022(2):143-147,181. |
| |
| Authors: | Zhao Shouyao Lu Huishan Wang Fujie LI Pei Wang Ning |
| |
| Institution: | (School of Mechanical Engineering,North China University,Taiyuan,030051,China) |
| |
| Abstract: | The behavior of laying hens in the growing period can reflect their health status and adaptability to the environment. In order to quickly identify the behavior of laying hens, this paper proposes a behavior recognition method based on the contour features of laying hens. Firstly, the top view image of layers was obtained, and the layer image was preprocessed. The contour of layers was obtained and fitted by automatic threshold segmentation, and the geometric features of the fitted contour were extracted. Then, the features were arranged and combined to obtain four feature combinations, which were trained with Extreme Learning Machine(ELM) to obtain the best feature combination. Finally, the behavior recognition model of layers was established. The average recognition time of this method is 1.54 s, and the recognition rate was high. The recognition rates of feeding, standing, lying, and dressing of a single layer were 97.00%, 94.46%, 91.50%, and 86.64%, respectively. |
| |
| Keywords: | behavior recognition feature extraction image processing extreme learning machine |
| 本文献已被 维普 等数据库收录！ |
|
