Gradient learning-based method for amplification of super-resolution image

Abstract

The embodiment of the invention discloses a gradient learning-based method for amplification of a super-resolution image. The method comprises: inputting a to-be-amplified image in an RGB color mode; converting the image in the RGB color mode into one in a YUV color mode; carrying out channel splitting processing on the image in the YUV color mode so as to obtain a brightness channel and a hue-saturation channel by splitting; amplifying the split hue-saturation channel by needed times according to an interpolation method; on the basis of gradient learning, amplifying the split brightness channel by needed times; carrying out space coupling on the amplified hue-saturation channel and the amplified brightness channel and converting the coupled space into RGB space and obtaining an image that is amplified by the times and in a RGB color mode; and outputting the amplified image in the RGB color mode. According to the embodiment of the invention, gradient of a brightness domain image is obtained; and high frequency information of the image is definitely reserved, thereby avoiding high frequency information loss.

Description

technical field [0001] The invention relates to the technical field of digital home, in particular to a super-resolution image enlargement method based on gradient learning. Background technique [0002] In a large number of electronic imaging applications, it is often desirable to obtain high-resolution (abbreviated as HR) images. High resolution means a high density of pixels in an image, providing more detail that is essential in many practical applications. For example, high-resolution medical images are very helpful for doctors to make correct diagnoses; using high-resolution satellite images, it is easy to distinguish similar objects from similar objects; if high-resolution images can be provided, computer vision The performance of pattern recognition will be greatly improved. [0003] Since the 1970s, charge-coupled devices (CCDs), CMOS image sensors have been widely used to capture digital images. Although these sensors are suitable for most imaging applications, ...

AI Technical Summary

Problems solved by technology

Because this party needs to use external information, the calculation of the algorithm is very large. At the same time, there is a situation where the "distance" between the image in the library and the image to be enlarged is small, but the visual difference is large, which reduces the quality of the enlarged image. the quality of

Images