A fast iso denoising method and system based on lifting wavelet transform

Abstract

The invention discloses a fast ISO denoising method and system based on lifting wavelet transform, relates to image processing technology, and aims to propose a fast ISO denoising algorithm capable of simultaneously retaining low-contrast information and high-contrast edge information of an image. Technical points of the present invention: decompose the image into a single channel to obtain the Y channel image, U channel image and V channel image; carry out n-layer linear lifting wavelet transform on the Y channel image; pair the results of the wavelet transform of the 3rd to n layers one by one Denoise the wavelet detail image in the image; wavelet inverse transform obtains the denoised approximate image in the first layer wavelet transform result; denoise the approximate image in the first layer wavelet transform result again; wavelet inverse transform obtains the original space Denoising the Y-channel image in the original space; performing wavelet transformation and denoising on the U-channel image and V-channel image respectively; and obtaining the denoising result of the three-channel image.

Description

technical field

[0001] The invention relates to image processing technology, in particular to a fast ISO denoising method. Background technique

[0002] In recent years, with the digitization of cameras and the use of mobile phones, an era of photography for all has come. People can take pictures at any time, in any scene, and with any focus.

[0003] In low-brightness or fast-changing scenes, the ISO speed of the image sensor is usually increased to increase the photosensitive speed, so that the camera can obtain more light in a short time. Increasing the speed of ISO is usually achieved by increasing the gain of the electrical signal and lowering the photosensitive threshold. When the speed of the sensor is increased, more random noise is often introduced. If the ISO speed is increased above 400, the random noise will gradually become more obvious.

[0004] The noise brought by high ISO greatly reduces the quality of the image, and because of the increase of useless noi...

Examples

Embodiment Construction

[0039] All features disclosed in this specification, or steps in all methods or processes disclosed, may be combined in any manner, except for mutually exclusive features and / or steps.

[0040] Any feature disclosed in this specification, unless specifically stated, can be replaced by other alternative features that are equivalent or have similar purposes. That is, unless expressly stated otherwise, each feature is one example only of a series of equivalent or similar features.

[0041] Such as figure 1 , the present invention includes:

[0042] Step 1: Convert the RGB format image to YUV420P format. Perform single-channel decomposition on the converted YUV format image: use the Y value of the pixel as the pixel value of the corresponding pixel in the Y channel image, and so on to obtain the U channel image and the V channel image.

[0043] Step 2: Perform n-layer linear lifting wavelet transform on the Y channel image; n is 5 in this preferred embodiment, and n can also be a...

technical field

[0001] The invention relates to image processing technology, in particular to a fast ISO denoising method. Background technique

[0002] In recent years, with the digitization of cameras and the use of mobile phones, an era of photography for all has come. People can take pictures at any time, in any scene, and with any focus.

[0003] In low-brightness or fast-changing scenes, the ISO speed of the image sensor is usually increased to increase the photosensitive speed, so that the camera can obtain more light in a short time. Increasing the speed of ISO is usually achieved by increasing the gain of the electrical signal and lowering the photosensitive threshold. When the speed of the sensor is increased, more random noise is often introduced. If the ISO speed is increased above 400, the random noise will gradually become more obvious.

[0004] The noise brought by high ISO greatly reduces the quality of the image, and because of the increase of useless noi...

Examples

Embodiment Construction

[0039] All features disclosed in this specification, or steps in all methods or processes disclosed, may be combined in any manner, except for mutually exclusive features and / or steps.

[0040] Any feature disclosed in this specification, unless specifically stated, can be replaced by other alternative features that are equivalent or have similar purposes. That is, unless expressly stated otherwise, each feature is one example only of a series of equivalent or similar features.

[0041] Such as figure 1 , the present invention includes:

[0042] Step 1: Convert the RGB format image to YUV420P format. Perform single-channel decomposition on the converted YUV format image: use the Y value of the pixel as the pixel value of the corresponding pixel in the Y channel image, and so on to obtain the U channel image and the V channel image.

[0043] Step 2: Perform n-layer linear lifting wavelet transform on the Y channel image; n is 5 in this preferred embodiment, and n can also be a...

Images