Image background blurring method and mobile terminal

Abstract

The invention discloses an image background blurring method. The method comprises the steps that a first image and a second image are subjected to downsampling processing separately, and a first downsampling image and a second downsampling image are obtained; a depth image is obtained according to the first downsampling image and the second downsampling image; the far point depth and the near point depth are obtained through calculation according to the depth image; the blur circle radius corresponding to each pixel is calculated; the first downsampling image is blurred according to the blur circle radius of each pixel, and the blurred first downsampling image is obtained; the blurred first downsampling image is subjected to upsampling to have the same size as the first image and serves asa first blurred image; the first blurred image and the corresponding first image are blended, and the first image obtained after a background is blurred is obtained. The invention discloses a mobileterminal used for executing the blurring method.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a method for blurring an image background and a mobile terminal. Background technique [0002] Blurring the background of an image is well known and used by many photography enthusiasts because it can quickly highlight the subject. Taking the current popular mobile phone with dual cameras as an example, one of its mainstream functional applications is to use the dual cameras to generate stereo vision, obtain the depth information of the image through binocular ranging, and then use the depth information to blur the background of the image . [0003] In order to achieve the ideal processing effect, the background blur of the image is generally realized by image convolution. However, the convolution operation itself is time-consuming, especially when the background blur using the convolution algorithm is applied to the mobile terminal. It is inevitable that there are pro...

AI Technical Summary

Problems solved by technology

[0003] In order to achieve the ideal processing effect, the background blur of the image is generally realized by image convolution. However, the convolution operation itself is time-consuming, especially when the background blur using the convolution algorithm is applied to the mobile terminal. It is inevitable that there will be problems such as cumbersome operation, large amount of calculation, and large memory consumption.

Therefore, most of the image background blurring solutions commonly used in mobile terminals at present use technologies similar to fast Gaussian blur or anti-Gaussian blur to reduce the complexity of calculation and achieve processing acceleration, but this method cannot be used for point light sources. It is processed separately, so the processed result will be far from the actual optical effect

Images