BP神经网络在茶叶病害多分类识别中的应用研究

Application of BP neural network in tea disease classification and recognition

随着物联网技术的不断发展,农业逐渐走向网络信息化,而利用物联网技术对农作物进行病害识别已经成为当前的一大研究领域。为能科学、精准的识别农作物的病害,本研究利用计算机图像处理技术和BP神经网络算法,以茶叶病害为研究对象,运用图像分割法对病斑图像来进行分割,通过提取经过图像处理后的病害茶叶的形状特征并分类,再利用MATLAB实现BP神经网络模型进行分类检测,最终得到非常高的识别率。

With the continuous development of the Internet of Things,agriculture has entered the age of network and information,and the application of the Internet of Things in the identification of crop diseases has become a major research field. In order to scientifically and accurately identify crop diseases,we adopted computer image processing technology and BP neural network algorithm,took tea diseases as the research object,used image segmentation method to segment the disease spot image,extracted and classified the shape characteristics of the diseased tea leaves,and used MATLAB to realize classification and detection by BP neural network model. Finally,a very high recognition rate was achieved.