The invention relates to a target recognition method based on man-machine fusion, belongs to the field of remote sensing image processing and analysis, and solves the problem of target recognition andclassification of large-format multi-target remote sensing images. The method comprises the steps of obtaining an eye movement signal when an image analyst performs image analysis, generating an eyemovement heat map corresponding to an image, determining a candidate target area in the image according to a set heat threshold, and cutting the image to extract a candidate target; performing featureextraction on the candidate target by using a pre-trained convolutional neural network; and performing target identification and classification according to the features of the candidate targets. According to the method, the accurate and efficient target search capability of human eyes and the powerful target classification capability of the convolutional neural network are combined, large-formatvisible light remote sensing image target recognition is achieved, and the method can be extensively applied to infrared, hyperspectral and synthetic aperture radar remote sensing image target recognition.