Left and right view feature point matching processing method, terminal and system in binocular ranging

Abstract

The invention relates to a left and right view feature point matching processing method, a terminal and a system in binocular ranging, and belongs to the technical field of binocular ranging. The method comprises the steps of performing feature point extraction according to left and right views of a target object, and preliminary feature point matching is performed by using an Euclidean distance algorithm; screening the feature points preliminarily matched from the left view and the right view by using a first preset formula; and obtaining a matching degree value interval range according to the screened feature points of the left view and the right view, and performing strong matching extraction on the selected matching point pair according to the minimum value of the matching degree valueinterval range and a preset empirical constant. According to the method, the matching precision of the left and right view feature points in binocular ranging can be improved.

Description

technical field [0001] The present application belongs to the technical field of binocular distance measurement, and specifically relates to a matching processing method, terminal and system for left and right view feature points in binocular distance measurement. Background technique [0002] Palletizing safety accidents in hazardous chemical warehouses occur frequently, so strengthening the supervision of hazardous chemical palletizing is more and more worthy of attention. In related technologies, based on binocular surveillance video information, vision is used to monitor distance exceeding the standard, forming an intelligent and automatic supervision method. [0003] see figure 1 , figure 1 It is the principle diagram of traditional binocular ranging. figure 1 Among them, f is the focal length of the camera; O R , O T are the optical centers of the two cameras; B is the baseline, that is, the distance between the optical centers of the two cameras, Z is the distanc...

AI Technical Summary

Problems solved by technology

see figure 2 , figure 2 It is the left and right view matching point information map obtained by the traditional SURF matching algorithm in the night vision environment, through figure 2 It can be seen that although the traditional SURF matching algorithm can detect matching information, it will generate a large amount of matching point information, and the number of matching points in the left and right views is different, including a large amount of incorrect matching information, too much matching information, resulting in feature point matching information The accuracy is not high, which will adversely affect the subsequent ranging results

Images