The invention relates to a crop disease and pest damage detection method and aims to solve the problems of low intelligent detection accuracy of crop disease and pest damage and large classification precision fluctuation at present. The method comprises the steps of preprocessing a visible light crop disease and pest damage image with a definite tag firstly, then establishing a convolutional neural network of a 18-layer network structure, performing convolutional neural network training by using a sample of a visible light crop disease and pest damage image training set to obtain a trained convolutional neural network, and extracting features of the visible light crop disease and pest damage image with the definite tag; training a support vector machine model by utilizing the extracted features and the classification tag of the visible light crop disease and pest damage image corresponding to the features; and extracting features of a newly uploaded visible light crop disease and pest damage image, and classifying the extracted features by using the trained support vector machine model. The method is suitable for the crop disease and pest damage image detection.