Image field depth drawing method and device

Abstract

The invention discloses an image field depth drawing method and device, and the method comprises the steps: carrying out the drawing of an original image in a preset mode, and generating a full-focusing image and a depth image; carrying out the highlight effect compensation for the full-focusing image, carrying out the layer-by-layer downsampling, and obtaining a plurality of scale images; calculating the radius of a dispersion circle of each pixel according to the depth image and a focusing position, and determining a diffusion kernel corresponding to each pixel on the corresponding scale image after the presetting of a bit order; carrying out the diffusion of each pixel in a diffusion radius range corresponding to the corresponding diffusion kernel according to the diffusion kernel corresponding to each pixel on the corresponding scale image after the presetting of the bit order; respectively carrying out the upward resampling of the scale images which are diffused or not diffused and the full-focusing image after highlight effect compensation, and obtaining a plurality of reconstructed images; carrying out the fusion of the plurality of reconstructed images through a preset weight coefficient, and carrying out the inversed highlight effect compensation. According to the above scheme, the method can simplify the calculation of image field depth drawing, and increases the field depth drawing speed.

Description

technical field [0001] The present invention relates to the technical field of image processing, in particular to an image depth-of-field rendering method and device. Background technique [0002] The depth of field effect refers to the object distance range within which the camera lens can form a clear image of the object. Outside the object distance range, including the foreground and background, the camera lens forms a blurred image of the object. The depth-of-field effect is an important optical property of a camera lens used to emphasize the subject being photographed to enhance the realism of the image. At present, the technology of rendering the depth of field effect has become an important research direction in computer graphics. [0003] A depth-of-field rendering method in the prior art is the forward mapping method, which uses the color and depth information in the source image to calculate the size of the circle of confusion of pixels at different positions in t...

AI Technical Summary

Problems solved by technology

[0004] However, this method needs to calculate the corresponding diffusion radius for each pixel in the source image and perform the diffusion operation when drawing the depth of field, so there is a problem of high computational complexity

Images