Face super-resolution reconstruction method based on deep learning

Abstract

The invention discloses a face super-resolution reconstruction method based on deep learning, and aims to utilize the deep learning technology to train low resolution face data, thus obtaining a mapping function from the low resolution face to the high resolution face; the technical keys comprise the following steps: 1, extracting key points of a training face data set; 2, calculating a face angleaccording to the extracted key points, and selecting a relatively right face image; 3, correcting the relatively right face image; 4, segmenting the corrected face image into the left eyebrow, the left eye, the right eyebrow, the right eye, the nose and the mouth, and respectively training said parts; 5, super-resolution processing the eyebrow, eye, nose and mouth images, super-resolution processing the face image, and combining said images so as to obtain the final super-resolution face image. The face super-resolution reconstruction method based on deep learning can effectively improve theobtained face image quality without changing the imaging system hardware equipment.

Description

technical field [0001] The invention belongs to the field of image processing, in particular to a face super-resolution reconstruction method based on deep learning. Background technique [0002] Image super resolution (super resolution, SR) is the process of obtaining a high-resolution image from a low-resolution image. This technology is mainly used to enhance the spatial resolution of the image, which can break through the limitations of the original system imaging hardware conditions. Re-acquired high-resolution images have the characteristics of higher resolution, more detailed information, and higher quality picture quality, and are currently one of the most effective and lowest-cost ways to obtain high-precision images. [0003] In the field of video surveillance, because the camera has a wide field of view and is far away from the face, the detected face is often very small. At the same time, due to the limitations of imaging conditions and imaging methods, the imagi...

AI Technical Summary

Problems solved by technology

[0006] (1) SR is an inverse problem. For a low-resolution image, there may be many different high-resolution images corresponding to it

[0007] (2) The low-quality images obtained in real multimedia applications are often complex degraded images with multiple degraded factors, such as low resolution, out-of-focus blur, motion blur, compression distortion and noise, etc.

Images