Fecha de publicación: 20/10/2023 Fuente: WIPO "processing tomato" The invention relates to a tomato leaf disease detection method based on deep learning, and belongs to the technical field of plant disease detection. The method comprises the following steps: S1, collecting common tomato leaf disease images, zooming and splicing the images, and expanding a data set by using an image processing method; s2, marking and classifying according to disease types of tomatoes, dividing a training set and a verification set, and finally obtaining a tomato leaf disease image data set; s3, constructing a small target disease identification model based on an improved yolov7-tiny network structure; s4, setting initial parameters of the model, training the constructed model by using the data set to obtain optimal model parameters, and storing the optimal model; and S5, identifying a to-be-detected image by using the trained model, and detecting a specific position and a name of a disease in the image. According to the invention, the detection effect of small target diseases can be rapidly and effectively improved, and the balance between precision and speed is achieved.