Dynamic video image clarity enhancement method and device

Abstract

The invention belongs to the image processing technical field and discloses a dynamic video image definition enhancement method and device. The method comprises the following steps that: the YUV data of current pixels are acquired, normalization processing is performed on the YUV data of the current pixels; neighborhood fuzzification is performed on Y components in the normalized data, and the difference values of the Y components before and after the fuzzification are calculated, and the gain coefficients of the Y components are calculated by using the difference values of the Y components; definition enhancement is carried out based on the Y components, the difference values of the Y components and the gain coefficients of the Y components; the RGB data of the current pixels are calculated by using the enhanced Y components and the UV components of the current pixels, and the RGB data of the current pixels are outputted. According to the method and device provided by the technical schemes of the invention, the color differences of different objects in an image are fast increased, so that efficient resolution enhancement can be realized, and continuous processing requirements of dynamic video images can be satisfied.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a method and device for enhancing the definition of a dynamic video image. Background technique [0002] Photography and camera equipment can help people record images so that they can be viewed at any time and place, but limited by the equipment and the ability of the photographer, the picture quality presented by many image materials is not ideal, and it is difficult to meet the needs of users. In the case of high cost of time and manpower and material resources for reshooting, certain technical means are usually chosen to enhance the clarity of the image. [0003] Traditional sharpness enhancement algorithms are often aimed at specific industries or working environments, such as pattern recognition in computers, medical X-ray imaging, meteorological imaging, etc. These only need to process individual static images, and the real-time requirements for processing are not...

AI Technical Summary

Problems solved by technology

[0003] Traditional sharpness enhancement algorithms are often aimed at specific industries or working environments, such as pattern recognition in computers, medical X-ray imaging, meteorological imaging, etc. These only need to process individual static images, and the real-time requirements for processing are not high. The processing capacity is usually large, which cannot meet the efficiency and performance requirements of continuous processing of dynamic images

In addition, the image enhancement methods in the prior art often only perform one-sided enhancement for a specific requirement, such as enhancing brightness, enhancing contrast, enhancing chroma, etc. Although the enhancement range is relatively large, the algorithm is generally relatively single. To enhance multiple parameters of the image, multiple algorithms need to be run separately, which is too large in calculation and poor in real-time performance

Images