Segmentation of Rapeseed Color Drone Images Using K-Means Clustering

文献类型： 外文期刊

第一作者： Yang, Kang

作者： Yang, Kang;Liu, Changhua;Wu, Xiaoming;Li, Hao

作者机构：

关键词： Drone rapeseed image; K-means clustering algorithm; Color segmentation algorithm; Image segmentation

期刊名称：PROCEEDINGS OF THE THIRD INTERNATIONAL CONFERENCE ON COMPUTER SCIENCE AND APPLICATION ENGINEERING (CSAE2019)

ISSN：

年卷期： 2019 年

页码：

收录情况： SCI

摘要： Aiming at the problem of unsatisfactory segmentation effect of aerial image of rapeseed flowers in the process of florescence recognition. This paper proposes a method combining K-Means algorithm and color segmentation algorithm to segment rapeseed image. Firstly, the K-Means algorithm is used to first process the rapeseed image in Lab space. Then, the clustering results were processed once again in HSV space using color segmentation algorithm. Finally, the segmented rapeseed was subjected to morphological treatment to complete the effective segmentation of rapeseed and rapeseed flowers. Sixty different aerial rapeseed images were selected for segmentation experiments. The results show that this method can not only segment rapeseed well, but also effectively avoid the influence of illumination. The results of this experiment can provide reference for the later study of the flowering period of rapeseed.

分类号：

  • 相关文献
作者其他论文 更多>>

微信小程序