High-angular-resolution light filed obtaining device and image generation method

Abstract

The invention provides a high-angular-resolution light filed obtaining device and an image generation method. The high-angular-resolution light filed obtaining device in which a light coding aperture is combined with a micro lens array is constructed to solve the problem that resolution tradeoff causes low spatial resolution or image aliasing leads to decrease of the quality of a light field image. The high-angular-resolution light filed obtaining device comprises a main lens, the coding aperture, the micro lens array and a sensor which are successively arranged along a light transmission path. On the premise that the positional resolution of the light field is not increased, depth information in light field data is used to improve the angular resolution of the light field. The high-angular-resolution light filed obtaining device has the advantages of being high in the angular resolution, high in the signal to noise ratio, high in the passing rate and capable of collecting light-field videos. The light-field image generation method using the compressed sensing theory has the advantages of being high in the angular resolution and capable of removing image aliasing.

Description

technical field [0001] The invention relates to the fields of computer vision, computational photography and optical engineering. Background technique [0002] A light field is a parametric representation of a 4D optical radiation field in space that contains both position and angle information. The theory of light field imaging has brought revolutionary progress to the field of digital imaging. Unlike traditional digital cameras that capture scene light through standard optical lens groups and digital signal sensors, light field cameras use a novel optical system to map the light in the scene to the sensor pixels in a given relationship. Based on the collected light field position information and angle information, a novel image is generated through calculation. However, compared with traditional imaging techniques, light field imaging techniques also have certain limitations. In the case of a certain number of sensor pixels, existing light field acquisition methods face...

AI Technical Summary

Problems solved by technology

It uses the framework of variational inequality to achieve simultaneous improvement of the angular resolution and position resolution of the light field, but this method requires sub-pixel-level precision scene geometric structure information as a priori, which is difficult to achieve in reality

Therefore, the existing methods to improve the position resolution cannot solve the problem of poor light field image quality and need to be improved

Images