The invention is suitable for the technical field of image processing, and provides a multi-frame noise reduction method and device based on deep learning, and the method comprises the steps: obtaining a first preset number of frames of original images; selecting one frame of original image in the first preset number of frames of original images as a reference frame of image; performing preliminary spatial domain noise reduction on the original images of the first preset number of frames through a spatial domain convolutional neural network; aligning each frame of other original image in the first preset number of frames of original images with the reference frame of image through a motion compensation convolutional neural network to align all the original images in the first preset numberof frames of original images; and splicing all the aligned original images in the first preset number of frames of original images, and performing convolution processing according to a time-space domain convolutional neural network to generate a denoised target image. According to the invention, the noise intensity contained in the image shot by the intelligent terminal is reduced, and the imagequality is improved.