Lettuce multispectral image foreground segmentation method

The invention relates to a lettuce multispectral image foreground segmentation method. Compared with the prior art, the defects that deviation exists between channels of a multi-lens multispectral camera and a lettuce foreground area cannot be accurately extracted by a traditional method are overcome. The method comprises the following steps: obtaining and preprocessing a lettuce multispectral image; performing row edge extraction operation; registering the multispectral image of the lettuce; establishing an image segmentation model; training an image segmentation model; obtaining and processing an image to be segmented; and obtaining a to-be-segmented image result. According to the invention, accurate registration of each channel can be carried out on the field lettuce multispectral image, foreground segmentation is realized, the registration time of a single image is 0.92 s, and the registration accuracy reaches 99%. 本发明涉及一种生菜多光谱图像前景分割方法，与现有技术相比解决了多镜头多光谱相机各通道之间存在偏差以及传统方法无法精确提取生菜前景区域的缺陷。本发明包括以下步骤：生菜多光谱图像的获取和