Fecha de publicación: 15/11/2019 Fuente: Wipo tomato The invention relates to the technical field of image recognition, in particular to an intelligent analysis method and an intelligent analysis device for tomato growth states, and the method comprisesthe steps: building a tomato growth analysis model in advance, and the tomato growth analysis model comprising a linear regression model and a convolutional neural network model; acquiring a tomato growth image, and respectively importing the tomato growth image into the linear regression model and the convolutional neural network model; analyzing and calculating the tomato growth image by the linear regression model to derive the current growth period growth index of the tomato; and the convolutional neural network model analyzing and calculating the tomato growth image to derive a current fruit leaf growth index of the tomato. The method has the beneficial effects that a linear regression model is adopted for qualitative analysis of the tomato growth image, a convolutional neural network model is adopted for quantitative analysis of the tomato growth image, and different models are adopted for qualitative analysis and quantitative analysis, so that the accuracy is improved, and theoperation rate is increased.