Pitaya yield prediction method based on machine learning

The invention discloses a pitaya yield prediction method based on machine learning. The pitaya yield prediction method comprises the following steps: (1) unmanned aerial vehicle remote sensing: shooting a pitaya base of a batch by an unmanned aerial vehicle to obtain a remote sensing image; (2) remote sensing image analysis: performing image processing and image recognition on the remote sensing image; (3) identifying and counting the pitaya: counting the number of the pitaya in the remote sensing image according to the remote sensing image in the step (2); (4) meteorological data feature calculation: analyzing through machine learning according to the meteorological factors in the growth period of the pitaya of the batch, obtaining a yield influence coefficient, and obtaining a predicted number and a predicted average weight through calculation; (5) historical yield calibration: according to the historical yield, performing calibration on the predicted yield to obtain a number calibration coefficient and an