Surface Classification Method Based on Background Noise Ratio of Single Photon LiDAR

Abstract

The invention discloses a ground surface classification method based on the background noise rate of single-photon laser radar. First, the expression of the photon reflection noise rate of the water surface is proposed according to the mirror reflection theory, and then the background is established by combining system parameters, environmental parameters and target characteristic parameters. The noise rate model provides the mathematical expressions of the land background noise rate and the water body background noise rate respectively, and finally calculates the threshold value of the surface classification noise rate. According to the significant difference between the background noise rate of land and water bodies, the statistical noise rate of the original point cloud data of lidar can be compared with the noise rate threshold to judge the surface type. This classification method does not rely on digital topographic maps or high-resolution remote sensing images used in traditional methods, and the auxiliary data used are easy to obtain, fast and efficient, and can achieve high-precision classification of surface types in coastal areas. The method is applied to MABLE original point cloud data, and the classification effect is excellent.

Description

technical field [0001] The invention relates to the technical field of surface classification, in particular to a surface classification method based on the single-photon laser radar background noise rate. Background technique [0002] Due to the higher detection sensitivity and higher repetition rate of photon counters and micropulse lasers, photon counting radars can obtain denser photon point cloud data than traditional full waveform radars. Therefore, in order to make more precise observations of the Earth's surface, NASA launched ICESat-2 (Ice, Cloud, and land Elevation Satellite-2) at the end of 2018 and equipped it with a photon counting radar. However, since photon counting radar is very sensitive to both signal photons and noise photons mainly reflected from the background light of the sun, this makes the original single-photon point cloud data have a lot of noise. Therefore, how to distinguish signal photons from the original point cloud data is a problem. The key...

AI Technical Summary

Problems solved by technology

[0005] Existing methods need to extract signal photons and compare them with high-resolution images for classification, NLCD auxiliary data is essential; and the noise rate threshold for water ice classification based on the background noise rate in the original point cloud data is based on experience At the same time, the current theoretical formula for the surface photon reflection noise rate is based on the assumption that the earth's surface is a Lambertian body, but this assumption is not applicable to the water surface, and there is no complete information on the theoretical formula for the water surface photon reflection noise rate.

[0006] It can be seen that there are technical problems of low accuracy and low efficiency in the prior art

Images