The invention relates to application of deep learning in the technical field of computer vision. The invention particularly relates to a fan blade defect intelligent detection method based on a double-spectrum image. According to the method, the image segmentation technology is utilized to realize segmentation of the blade area, background removal is realized, the identification efficiency and accuracy are improved, the infrared image and the visible light image are fused through the image processing technology, and then the deep learning technology is utilized to perform defect identificationon a large number of fan blade images. By accumulating a large amount of blade damage defect data and via techniques through big data, development rules of various damage defects in a large amount ofblade damage data are mined, a mathematical model is constructed, prediction and evaluation of the blade defects and the health state are finally achieved in combination with the machine automatic learning technology, the image processing technology and the double-light-source imaging technology are fused, the utilization rate of image information is effectively increased, and image feature information is obviously highlighted.