A Monocular Topology Matching 3D Imaging Method in Time Domain

Abstract

A monocular time-domain topology matching three-dimensional imaging method, which collects A and B frames of images for the target, wherein the A and B frames of images have overlapping image parts; perform feature point extraction and feature description on the A and B frames of images respectively; use Two-way feature matching performs feature matching on A and B frame images to obtain feature point pairs; according to the matched feature point pairs, constructs a topology structure for A and B frame images in turn; performs topology matching on A and B frame images, and performs B frame image Geometric correction, to obtain the overlapping areas of A and B frame image pixels one by one; super-resolution 3D reconstruction is performed on the overlapping areas of A and B frame images to achieve 3D imaging of the target. This method does not require high motion stability of the carrying platform , good applicability.

Description

technical field [0001] The invention relates to the technical field of three-dimensional imaging, in particular to a three-dimensional imaging method based on monocular topology matching based on time domain information. Background technique [0002] Range-gated 3D imaging is a 3D imaging technology with active lighting, which can be widely used in the fields of vehicle autonomous navigation and 3D scene reconstruction. The super-resolution 3D imaging in range gating utilizes the echo broadening effect to construct a range energy envelope that satisfies a specific geometric shape, and obtains the 3D information of the target through the inversion of two images. Long-range three-dimensional active imaging with superresolution depth mapping[J]. Optics Letters, 2007, 32 (21): 3146-3148.) and triangular distance energy envelope (Wang Xinwei, Li Youfu, Zhou Yan.Triangular-range-intensity profile spatial -correlation method for 3D supper-resolution range-gated imaging[J].Applied ...

AI Technical Summary

Problems solved by technology

As shown in Figure 1, the target in the two images collected under the condition of a moving platform or moving target will have position changes and distortions. During the reconstruction process, the range energy envelope constructed by the echo broadening effect cannot act on the same target position, and the wrong three-dimensional information of the target can be obtained

Therefore, for super-resolution 3D imaging under the condition of moving platform or moving target, accurate 3D information of the target cannot be obtained, and 3D reconstruction cannot be realized.

Images