Method for eliminating high light sensitivity speed noise and digital image processing equipment

Abstract

The invention relates to the digital image processing technology, and discloses a method for eliminating high light sensitivity speed noise and digital image processing equipment. In the invention, multi-order multi-resolution decomposition is carried out on each color channel of the digital image with high ISO speed noises, the high frequency part decomposed in each order is filtered, and each color channel is rebuilt by inverse decomposition. By judging and eliminating high ISO speed noises under multiple resolutions instead of within a certain window range, the noise elimination effect is obviously improved; and moreover, each pixel adaptive window is free of dynamic adjustment, thus omitting the process for calculating the signal statistical properties in quantity, thereby greatly reducing the operation complexity of noise elimination and facilitating implementation in portable or embedded systems.

Description

technical field [0001] The invention relates to digital image processing technology, in particular to noise elimination technology in digital image processing. Background technique [0002] With the continuous improvement of charge coupled device (Charge Coupled Device, referred to as "CCD") and complementary metal oxide semiconductor (Complementary Metal Oxide Semiconductor, referred to as "CMOS") digital image sensor technology, digital photography is gradually replacing traditional film photography . This trend is not only reflected in the field of professional journalists and photographers, but also in the daily life of the general public. This trend has promoted digital camera manufacturers to promote some classic technologies in traditional film photography, such as high film speed (or ISO speed, ISO is the English abbreviation of International Organization for Standardization), to digital photography. middle. [0003] In traditional film photography, the ISO standa...

AI Technical Summary

Problems solved by technology

[0005] (1) Impulsivity: In the case of low brightness and high ISO speed, some image sensor units only retain completely single color information due to oversaturation, thus forming some bright impulsive noise in the acquired image

[0006] (2) Color channel inhomogeneity: the usual digital image sensor is less sensitive to some color channels (such as blue) than other colors, so the compensation gain for such colors will be larger

[0016] (1) Dynamically adjusting the adaptive window of each pixel needs to calculate the signal variance multiple times, so the complexity is very high, especially not suitable for implementation in portable or embedded systems

[0017] (2) The implementation proves that the filtering effect of this method is poor in the case of large noise points and dense high ISO speed noise.

Images