A single-camera-based target space localization method

Abstract

The invention discloses a method for locating a target space based on a single camera, and belongs to the technical field of intelligent video monitoring. The method of the present invention aims at the problem that the existing technology of target space positioning based on the linear camera model does not consider the imaging error caused by the lens distortion. Based on the isotropic characteristics of the lens distortion, when using the ideal linear camera model for target space positioning, Combining GPS and other global positioning systems to obtain spatial positioning auxiliary information such as camera position, scene center reference point, and geodetic absolute coordinates of marking points, and compensate and correct the position of the target in the image, so that the actual geodetic absolute coordinates of the target can be obtained quickly and accurately coordinate. Compared with the prior art, the robustness, real-time performance and applicability of the method of the present invention are greatly improved.

Description

technical field [0001] The invention relates to a target space positioning method, in particular to a single camera-based target space positioning method, which belongs to the technical field of intelligent video monitoring. Background technique [0002] Video surveillance technology has been widely used in the past ten years. In daily life, various video surveillance systems can be seen in use everywhere, for example, video surveillance systems are usually installed in schools, roads, subways, railway stations, etc. played a major role. With the continuous development of society, people's demand for security continues to increase, especially with the maturity and perfection of computer, network, information and communication, audio and video codec and other technologies, making video surveillance systems continue to be digitized, networked, Intelligent direction development. [0003] In intelligent video surveillance, target location in the scene is a key technology, and...

AI Technical Summary

Problems solved by technology

Today's three-dimensional space positioning technology is mainly based on two camera models, one is a nonlinear model, which considers the nonlinear distortion of the camera lens, and uses the camera's internal and external parameter matrix to solve the problem, which has high computational complexity. It is difficult to apply in positioning. The other is the linear model, which does not consider the error caused by the camera lens distortion, directly uses the principle of small hole imaging, uses the comparison invariance in projective geometry, and forms a three-dimensional space to a two-dimensional model according to the projection ratio. The mapping of two-dimensional images can obtain the three-dimensional space coordinates corresponding to the two-dimensional image coordinates in a relatively short period of time. The disadvantage is that the error is large, and the positioning accuracy cannot be guaranteed.

Images