Full-reference light field image quality evaluation method based on multi-scale contourlet characteristics

Abstract

The invention relates to a full-reference light field image quality evaluation method based on multi-scale contourlet characteristics. A contour and a curve of a light field image surface have perception correlation; edges and contour details of scene structure changes in the light field image are effectively captured through multi-scale multi-direction contourlet transformation with good anisotropy, subjective perception of a human eye vision system to the light field image is fully considered, and good light field image quality evaluation performance is achieved. According to the embodimentof the invention, the evaluation result of the subjective perceptibility matched with the human eye vision can be obtained.

Description

technical field [0001] The invention relates to the field of image processing, more specifically, to a full-reference light field image quality evaluation method based on multi-scale contourlet features. Background technique [0002] With the development of optoelectronic technology, light-field imaging, as a technique that can capture richer visual information from a single image, has attracted extensive attention from academia and industry, and has been widely used in the fields of computer vision and computer graphics. Applications, such as multi-view imaging, 3D reconstruction, full-focus imaging, depth estimation, virtual reality, augmented reality, etc. Different from traditional imaging, light field imaging acquires optical flow data at any point in space in any direction, and the resulting light field image can often better reflect the position and direction information in the real scene. [0003] However, light field images will inevitably introduce various distort...

AI Technical Summary

Problems solved by technology

There is a lack of quality evaluation methods suitable for light field image quality evaluation in the prior art

Images