Object dimensional positioning method and camera

Abstract

The invention provides an object dimensional positioning method and a camera. The method comprises the following steps of: setting at least two imaging sensors in parallel to obtain the at least two plane images of the same object; computing the value of the coordinate C of a third axle according to the values of x1 and x2 of the coordinates of the object in the two plane images, wherein the coordinates of the object in the two plane images are respectively to be (x1, y1) and (x2, y2), and the dimensional coordinate of the object is (A, B, C); averaging the C value on the time domain to obtain the averaged mean value C'; leading the mean value of the x1 and the x2 to multiply by the C' and divide by the focal distance to obtain A; leading the mean value of the y1 and the y2 to multiply bythe C' and divide by the focal distance to obtain B; and performing the dimensional positioning to the object according to the coordinate (A, B and C) of the object. The invention reduces the shake of the dimensional coordinate on the basis of invariable hardware cost, and keeps the reaction speed on the X direction and the Y direction at the same time.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a method for three-dimensional positioning of an object and a camera. Background technique [0002] In binocular vision, the general method for locating three-dimensional objects is: set the intersection line between the imaging plane of the image sensor and the horizontal plane as the X axis, the vertical line of the X axis on the imaging plane of the image sensor is the Y axis, and the vertical line of the imaging plane of the image sensor is Z axis. First calculate the two-dimensional coordinates of the target object in the two plane images, and then use the triangular relationship to calculate the midpoint of the common vertical line connecting the optical center of the two plane images and the object coordinates, that is, the projection line, which is the three-dimensional coordinates of the object. [0003] From the two-dimensional coordinates of the target object...

AI Technical Summary

Problems solved by technology

[0004] Due to the quantization error of the cmos (integrated on the semiconductor material of metal oxide) or CCD (integrated on the semiconductor single crystal material) sensor that obtains the image, even if the sub-pixel coordinate algorithm is used, in the case of a narrow baseline, it will still cause three-dimensional Coordinate Jitter

Images