An image filtering method and device thereof

Abstract

The present invention relates to the field of communication electronics technology, and in particular, to an image filtering method and its device. The method includes: determining whether a pixel is a boundary pixel, and if not, using the pixel as a center point; selecting a reference matrix of the center point in a preset manner; calculating the value of the area to be interpolated in the reference matrix ; The area to be interpolated is adjacent to the pixel center point; calculate the value of the pixel center point according to the value of the area to be interpolated, and replace the value of the pixel center point. It can be seen that the present invention denoises the bayer image, performs filtering without destroying the original noise characteristics of the bayer image, and performs pixel value interpolation, which increases the amount of pixel information of the bayer image. It can not only effectively remove noise, but also ensure Denoising is reliable and facilitates hardware implementation.

Description

technical field [0001] The invention relates to the technical field of communication electronics, in particular to an image filtering method and device thereof. Background technique [0002] The pictures inside the camera are raw pictures saved in bayer format and stored in .raw format, which is an unprocessed and uncompressed format. The bayer format is stored in three different channels: R, G, and B. The storage format is as follows: figure 1 shown. [0003] The image is collected and stored by the CMOS sensor, and the reception of image information will be hindered due to external interference and noise generated by optoelectronic equipment and circuits. Therefore, denoising processing is required, and the currently commonly used denoising methods are as follows: [0004] General denoising is based on grayscale images or RGB single-channel filtering operations, each pixel is a true representation of a pixel, and the amount of information that can be obtained for a baye...

AI Technical Summary

Problems solved by technology

When the image is converted from bayer format to RGB format, an interpolation algorithm is required. Image interpolation will destroy the characteristics of the image itself and generate new noise.

Before interpolation, there are still some algorithms that will destroy the structure of the image itself, increase and increase the noise of the image, and increase the difficulty of image denoising

[0006] The ordinary bayer image denoising has only 1 / 4 of the information of the gray image, and the pixel value can only be extracted interlaced in the pixel extraction. When the window is small, it is difficult to extract a large amount of information of the same channel, so It brings great difficulty to filtering and greatly reduces the accuracy of filtering

Images