Image de-noising algorithm based on Gaussian mixture model and adaptive threshold modeling

Abstract

In this paper, we conduct research on the image de-noising algorithm based on Gaussian mixture model and adaptive threshold modeling. Minimum spanning tree based on the local threshold value method, can make full use of the image information change itself, by defining a new local threshold method to calculate the regional difference and extracting, gives image a good segmentation and de-noising effect. Gaussian mixture model is a kind of important background modeling method. The method using multiple Gaussian background model is established through the background update constantly adjust background model of the components of the Gaussian distribution, and therefore has a certain ability to adapt. Under this assistance, we integrate the corresponding mode to propose the new image de-noising algorithm that obtains better performance.