Building relative distance estimation method based on near-infrared tri-spectral imaging

Abstract

The invention discloses a building relative distance estimation method based on near-infrared tri-spectral imaging. The method comprises the following steps: firstly, measuring the spectrum of sunlight in a foggy day and selecting three near-infrared bands from the spectrum; then, using the same near-infrared camera and the optical filters corresponding to the three near-infrared imaging wavebands for imaging the selected building target in the foggy weather to obtain three near-infrared images imaged in different wavebands in the same scene; respectively determining global atmospheric light values and extinction coefficients of the three near-infrared images; deducing a building relative distance estimation formula according to the three near-infrared images and the corresponding global atmospheric light values and extinction coefficients; and finally, estimating the relative distance of the building by using the derived distance estimation formula. According to the invention, the influence of the reflectivity of the material under different wave bands on distance estimation in the prior art is overcome, so that the estimation of the relative distance of the building is more accurate.

Description

technical field [0001] The invention belongs to the technical field of image visual distance estimation, and in particular relates to a method for estimating relative distances of buildings based on near-infrared three-spectrum imaging. Background technique [0002] Distance perception technology is widely used in national defense and civilian fields, such as automatic barriers based on distance measurement technology, unmanned driving, and robotic arm grabbing targets. Currently, there are two main types of distance perception based on the type of light source. The first is a distance perception method based on active light, such as a method based on triangulation ranging and time-of-flight (ToF). This distance perception method based on active light needs to provide a light source as a signal source, and is susceptible to interference from external conditions such as noise. , has certain requirements on the surface material of the measured target, and its performance is p...

AI Technical Summary

Problems solved by technology

The disadvantage of this method is: because this method relies on the distance measuring sensor, but the sensor signal is unstable in long-distance measurement, the distance measured by this method has an estimation error of about 7% within 2.5km, so this method is not applicable estimate at a longer distance

The shortcomings of this method are: the original distance estimation model established assumes that the reflectance of the same material at 905nm and 980nm bands is equal, but for some metal materials, the assumption is not true; and the wavelength selection is limited to two Specific near-infrared wavelengths, and other wavelengths cannot be used for imaging to estimate the distance, and there is a certain error in the estimation results, about 10%

Images