Object distance estimation method based on detection system

Abstract

The invention provides an object distance estimation method based on an imaging detection system, mainly solving the problems of great computation quantity and poor reliability of the traditional object distance estimation method. The estimation method comprises the following steps of: acquiring the azimuth angle sequence and the pitch angle sequence (alpha n, beta n) of an object by using an electro-optic theodolite; acquiring the space coordinate position (xn, yn, zn) of an imaging system through a GPS; acquiring the image sequence Img[n] of the object through a camera; determining the direction cosine of the object according to the azimuth angle and the pitch angle, matching the object images at two adjacent sampling moments n and n+1 to obtain the characteristic linearity Ln and L(n+1) and the object distance ratio rho=rn / r(n+1), and calculating the ratio coefficient H', the second-order coefficient D2, the first-order coefficient D1 and the constant coefficient D0 based on the characteristic linearity and the object distance ratio; constructing an object distance estimation equation by using the coefficients, and solving the object distance estimation equation to obtain the distance r(n+1) between the object and the imaging system at n+1. The invention has the advantages of distance estimation time saving and no defective solution and is applicable to the object distance estimation of a focus-variable imaging system.

Description

technical field [0001] The invention belongs to the technical field of imaging detection and relates to the estimation of the target distance by a passive tracking system, in particular to a method for estimating the distance between the target and the camera according to the imaging characteristics of the target, the spatial coordinates of the camera, the azimuth angle and the elevation angle of the target. Background technique [0002] The passive tracking system based on imaging detection has important theoretical significance and application value for the estimation of target distance. At present, when the target is imaged and detected, it is easy to obtain the relative distance change of the target, and then it can be inferred whether the distance between the target and the imaging system has increased or shortened. However, it is difficult to accurately estimate the distance between the target and the imaging system based on the target image features without prior info...

AI Technical Summary

Problems solved by technology

However, it is difficult to accurately estimate the distance between the target and the imaging system based on the target image features without prior information of the target (such as appearance size, or initial distance, etc.), which often requires complex iterations Calculation, real-time is not easy to guarantee, such as the method in the document "Tao Lin, Huang Shike, Zhang Tianxu. Optical flow based passive ranging technology for infrared targets. Computer and Digital Engineering, 2005, 33(11): 1-4."; some The method tries to estimate the target distance based on simple image features and regularized operations, such as the literature "Raghuveer Rao, Seungsin Lee. AVideo Processing Approach for Distance Estimation. IEEE ICASSP 2006, Part III: 1192-1195.", but it The distance estimation error is large, and it cannot adapt to the large non-roundness of the target edge; the literature "Fu Xiaoning, Liu Shangqian. Single-station passive ranging based on optoelectronic imaging, Optoelectronic Engineering, 2007, 34(5): 10-14 . "Proposed a distance estimation method that comprehensively utilizes target imaging features, camera space coordinates, target azimuth and pitch angles, it estimates the distance of the target to the camera by solving the quadratic equation

Although this method no longer needs the initial distance of the target, in practical applications, due to measurement errors and other reasons, the one-dimensional quartic equation is prone to ill-conditioned solutions, such as complex number solutions and negative value solutions, which limits its practicability

To sum up, the disadvantages of existing methods are large amount of calculation, poor real-time performance or poor practicability

Images