Corn seed approximate variety screening method, device and equipment and storage medium

The invention discloses a corn seed approximate variety screening method and device, equipment and a storage medium. The method comprises the following steps: acquiring a sample data set; the sample data set comprises fusion data of RGB images and multispectral images of a plurality of groups of corn seeds; inputting the sample data set into a constructed convolutional neural network model based on a multi-head self-attention mechanism for model training to obtain a trained corn seed approximate variety screening model; and inputting fusion data of the RGB image and the multispectral image of the to-be-detected corn seed into the corn seed approximate variety screening model to obtain a screening result of the corn seed approximate variety. According to the method, through deep learning and multispectral technologies, only the RGB images and the multispectral images of the corn seeds are needed, the approximate corn varieties can be rapidly and efficiently screened, compared with a traditional method, the met