Light field camera extrinsic parameter calibration device and method

Abstract

The invention provides a light field camera external parameter calibration device and method. According to the device and method, a color coding plate is arranged before the main lens of a light field camera and is fixed on an angle rotating table; the light field camera shoots the color coding plate arranged in different angles, corresponding relations of colors and angles of light emitted by the color coding plate in various angles are adopted as prior information; the light field camera is arranged at a plurality of locations and captures scenes; and external parameter calibration can be completed based on the priori information. With the device and method of the invention adopted, a plurality of external parameters of double planes in arbitrary attitudes can be calibrated, and the calibration does not depend on a fixed mode camera movement platform, and therefore, external parameter calibration is more accurate, and is not effected by errors caused by the inertia of the camera movement platform.

Description

technical field [0001] The invention relates to the fields of computer vision, computational photography and optical engineering, in particular to a color-coded plate-based light field camera external parameter calibration device and method, which can calibrate the rotation and translation relationship between multiple double-plane light fields. Background technique [0002] Light field imaging theory is a research hotspot in the field of computational photography. Based on this, foreign research institutions have designed a variety of single-view light field imaging systems with their own characteristics. Among them, light field cameras based on microlens arrays have become mainstream light field acquisition devices. However, limited by the imaging principle and sensor manufacturing process, the single-vision light field imaging system is not complete enough to sample the target light field signal, resulting in a limited field of view, insufficient depth of field, low positi...

AI Technical Summary

Problems solved by technology

On the basis of the simplified light field camera model, Xue et al. deduced the plane transformation relationship between multiple light field cameras, and realized high-quality multi-view light field stitching and fusion under the condition that the motion of the light field cameras is known. When the movement of the light field camera is unknown, the plane transformation between light field cameras cannot be realized

Therefore, in the case of unknown rotation angle and translation distance of the light field camera, the existing algorithm cannot obtain the extrinsic parameters of the light field camera

Images