Rice leaf age detection method based on computer vision and multi-feature fusion

The invention discloses a rice leaf age detection method based on computer vision and multi-feature fusion. The method comprises the following steps: acquiring an image of a rice plant to be detected; obtaining leaf age and multi-feature data of a plurality of rice plants; training the leaf age and multi-feature data of the rice plant through a plurality of machine learning models, and comparing and improving to obtain an optimal multi-feature fusion leaf age detection model; processing the to-be-detected rice plant image to obtain height, tiller number, total leaf number, main stem leaf number and spike number characteristic data of the to-be-detected rice plant; calculating the number of days from the sowing time; and forming a feature matrix according to multiple features of a to-be-detected rice plant, and inputting the feature matrix into the leaf age detection model to obtain a detection result. According to the method, multiple machine learning models are used for training rice feature data, and a mult