摘要
利用图像处理技术分析小麦图像的颜色特征,并提出叶片盖度(LCD)参数;通过小麦种植密度与不同参数之间的相关性分析,建立基于不同参数的预测模型。结果表明:以绿光标准化值(NDIG)和LCD结合的多元逐步回归模型优于其他单变量模型以及混合模型,可以作为小麦种植密度的定量估算模型。使用实测数据对模型进行验证,R^2为0.896 1,达到极显著水平。结果说明该方法可行,可为小麦生长前期种植密度的快速、精确检测提供依据。
The image color feature of wheat was analyzed by using image processing and the leaf cover degree (LCD) parameter was proposed. The prediction model was established based on different parameters through the correlation analysis between the wheat planting density and different parameters. The results showed that the multiple stepwise regression model with green standardized values (NDIG) and LCD was superior to other single model and hybrid model, which could be used as the quantitative estimation model of wheat planting density. Using the measured data to verify the model, R2 was 0. 896 1, reached extremely significant level. The method is feasible, and provides the basis for fast and accurate estimation of the early stage of the wheat planting density.
出处
《扬州大学学报(农业与生命科学版)》
CAS
北大核心
2017年第1期89-93,共5页
Journal of Yangzhou University:Agricultural and Life Science Edition
基金
江苏省农业三新工程项目(SXGC[2015]330)
江苏高校优势学科建设工程项目(2014-12)
关键词
小麦
种植密度
图像处理
颜色特征
叶片盖度
估算模型
wheat
planting density
image processing
color feature
leaf cover degree
estimation model