Multi-ocular image refocusing depth acquisition method and system

Abstract

The invention relates to a multi-ocular image refocusing depth acquisition method and system. The depth acquisition method comprises the steps that the optical center positions of multiple cameras arecalibrated; a scene light field is restored according to the image planes acquired by all the cameras and the corresponding optical center positions, and the distribution situation of the light in the space is recorded through a virtual plane; the positions of all the cameras are virtualized into synthetic aperture so that the light through the virtual plane is enabled to form the image in the image plane of the synthetic aperture; the image plane of the synthetic aperture is moved and the refocus images under the corresponding focal length are acquired; and the refocus images are extracted so that the depth of each point in the three-dimensional scene can be acquired. The scene light field is restored according to the image planes acquired by all the cameras and the optical center positions, the positions of all the cameras are virtualized into the synthetic aperture, the image plane of the synthetic aperture is moved to realize multi-ocular image refocusing, and the refocus images are extracted so that the error of high repetition texture area estimation can be reduced and the robustness of the high repetition texture area can be enhanced.

Description

technical field [0001] The invention relates to the technical field of computer image processing, in particular to a depth acquisition method and system for multi-purpose image refocusing. Background technique [0002] In computer vision systems, 3D scene information provides more possibilities for various computer vision applications such as image segmentation, target detection, and object tracking, and depth maps (Depthmap), as a common way to express 3D scene information, have been obtained. Wide range of applications. [0003] The gray value of each pixel in the depth image can be used to represent the distance of a point in the scene from the camera. Methods to acquire depth images can be divided into two categories: active depth sensing and image-based methods. [0004] For example, the Zcam depth camera developed by 3DV Systems uses an infrared pulse light source to send signals to the scene, and then uses an infrared sensor to detect the infrared light reflected ba...

AI Technical Summary

Problems solved by technology

[0011] In order to solve the above-mentioned problems in the prior art, that is, in order to solve the problem of low robustness in areas with high repetitive textures, the present invention provides a depth acquisition method and system for multi-eye image refocusing

Images