Method for measuring distance between objects by using projections of objects in video

Abstract

The invention provides a method for measuring the distance between objects by using projections of the objects in a video. The method comprises the following steps of S01, building a model, wherein apoint A is set as a vertex of a camera, a point O is set as an intersection point of the bottom surface of the camera, points B and C are set as two measured objects, projections of the points B and Cand the ground are OB and OC respectively, a point B' is the projection of the point B and the ground in a video picture, a point C' is the projection of the point C and the ground in the video picture, the point A is connected with the point O by a straight line, the point A, the point B and the point B' are connected by a straight line, the point A, the point C and the point C' are connected bya straight line, the point O, the point OB and the point B' are connected by a straight line, the point B is connected with the point B' by a straight line, the point B is connected with the point OBby a straight line, the point C is connected with the point OC by a straight line, and the point OB is connected with the point OC by a straight line, so that the model is built; and calculating thedistance between the OB and the OC through mathematical calculation. According to the method, data of two-dimensional coordinates of an original video picture are not used and the calculation error caused by lens distortion is eliminated.

Description

technical field [0001] The invention relates to a method for measuring distance, in particular to a method for measuring distance between objects by using object projection in video. Background technique [0002] There are currently two main methods for measuring the distance between objects in a video: one is to measure directly on-site, which is the most commonly used method, but if the object to be measured is not there, it cannot be measured; the other is to measure the length of objects on the same focal plane or near the focal plane, According to the scale conversion of the two-dimensional plane coordinate length in the picture, since the lens distortion is a nonlinear change, there is a certain systematic error in this method, and most of the cameras are looking down, so the focal plane is difficult to determine; the third is to measure the shooting location on the spot Multiple sets of three-dimensional coordinate data are associated with the two-dimensional coordina...

AI Technical Summary

Problems solved by technology

[0002] There are currently two main methods for measuring the distance between objects in a video: one is to measure directly on-site, which is the most commonly used method, but if the object to be measured is not there, it cannot be measured; the other is to measure the length of objects on the same focal plane or near the focal plane, According to the scale conversion of the two-dimensional plane coordinate length in the picture, since the lens distortion is a nonlinear change, there is a certain systematic error in this method, and most of the cameras are looking down, so the focal plane is difficult to determine; the third is to measure the shooting location on the spot Multiple sets of three-dimensional coordinate data are associated with the two-dimensional coordinate data in the screen, and the mapping relationship between the two-dimensional coordinates of the screen and the real three-dimensional coordinates is solved by using methods such as affine transformation, thus obtaining the three-dimensional coordinates of each unknown point, using The distance formula calculates the length. Although this method has a wide range of applications, the calculation process is complicated. Due to factors such as lens distortion, the three-dimensional coordinate data measured on the spot must include the measured target

Images