Inter-frame noise reduction method based on motion detection

Abstract

The invention discloses an inter-frame noise reduction method based on motion detection. According to the method, moving targets are extracted by a multi-Gaussian mixture background model method and an overlapping stationary area between two adjacent frames are found, then inter-frame accumulative filtering is performed on the area, and moving target areas and non-moving target areas in non-overlapping areas are replaced by background models established by an intra-frame filtering algorithm and the multi-Gaussian mixture background model method respectively. Meanwhile, the algorithm can also self-adaptively adjust the number of stack frames and has a multistage adjustable function. The innovative points reside in that moving target detection of images is performed firstly, then AND operation is performed on two successive frames of foreground images including the moving targets only, and the inter-frame filtering algorithm, the intra-frame filtering algorithm or a background model replacement algorithm is selected according to the result of AND operation so that the phenomena of edge virtual images, pseudo images and even lost of the moving targets caused by the conventional inter-frame filtering algorithm can be avoided, and the great noise reduction effect of the moving images can also be achieved.

Description

technical field [0001] The invention relates to a noise reduction processing technology of a video image, in particular to an inter-frame noise reduction method based on motion detection. Background technique [0002] In reality, digital video signals often introduce noise when they are acquired, transmitted, stored or reconstructed, and these noises can cause blurred images, making the images whitish or too dark. In order to improve the subjective visual effect of video sequences, improve video compression efficiency, save transmission bandwidth or perform subsequent video processing and analysis, it is necessary to denoise the noisy video signal. [0003] In video images, especially for static or slow-moving image sequences, there is generally a strong correlation between adjacent frames, while noise is always randomly distributed in video images and has no correlation, so the inter-frame The correlation of the signal and the non-correlation of the noise, first judge the ...

AI Technical Summary

Problems solved by technology

However, these methods also have their own technical shortcomings: 1. Mean filtering is to set the gray value of each pixel in the image to the average gray value of the pixels in the neighborhood of the point, but this method destroys the edges and edges of the image. The details make the image blurred; 2. The adaptive Wiener filtering method uses the correlation characteristics and spectral characteristics of a stationary random process to filter, so that the mean square error between the original image and the filtered image is minimized

This method can protect the edge and detail information of the image, but the amount of calculation is large, and it is not suitable for the situation where the noise is a non-stationary random process; 3. The median filter method is to set the gray value of each pixel in the image to the The median value of the gray value of all pixels in a certain neighborhood window

This method can protect edge information, but will cause blurred details; 4. In addition to filter transformation, frequency domain filtering also needs to perform Fourier forward transform and inverse transform, and the filtering time is longer than spatial domain filtering and time domain filtering; 5. Between frames Mean filtering is a weighted average of the pixels at the same position of adjacent frame images. This method has a better denoising effect due to consideration of the correlation between images between frames, but it will cause blurring and loss of moving objects, and is only suitable for stillness Image; 6. The non-local filtering algorithm and the three-dimensional block matching filtering method proposed in recent years are currently considered to have the best denoising effect, but the algorithm is complex and difficult to implement in hardware

Images