Urban wall multispectral imaging disease nondestructive detection method based on neural network

The invention discloses an urban wall multispectral imaging disease nondestructive detection method based on a neural network, and relates to the technical field of digital image processing, and the method comprises the following steps: obtaining urban wall multispectral data after black and white plate correction and registration; carrying out dimension reduction and denoising on the obtained multispectral data through a minimum noise separation method; fusing the low-dimensional images after dimension reduction and noise reduction, and sending the fused low-dimensional images into an established CNN model for training and testing; the processed to-be-detected city wall multispectral data are sent into the model for prediction, and a detection result is output and visually displayed; according to the invention, pixel-level classification detection is carried out on the urban wall diseases by using the trained convolutional neural network classification model with perceptual ability, and the prediction result