Photon point cloud denoising method and system based on laser radar channel line scanning characteristics

Abstract

The invention provides a photon point cloud denoising method and system based on laser radar channel line scanning characteristics. The method comprises the following steps: calculating a k density value of each photon point; removing the photon points of which the k density values are greater than a first threshold value; calculating the slope between each projection point and the nearest projection point; calculating the slope of a laser radar channel detection direction; performing interval division on the first residual photon points along the detection direction of the laser radar channel; calculating the length of each interval; calculating Euclidean distances from the first residual photon point corresponding to the minimum k density value in each interval to other first residual photon points; removing the first residual photon points of which the Euclidean distances are greater than a second threshold value in each interval; extracting an interval of which the interval length is greater than two times of a second threshold value; calculating the average Euclidean distance of each second residual photon point in each extraction interval; and removing the second residual photon points of which the average Euclidean distance is greater than a third threshold value in each extraction interval. According to the invention, signal-to-noise separation can be effectively realized.

Description

technical field [0001] The invention belongs to the technical field of laser radar, and in particular relates to a photon point cloud denoising method and system based on the characteristics of laser radar channel line scanning. Background technique [0002] Lidar uses lasers as the radiation source, which can accurately detect the distance, position, and movement trajectory of the target, and measure the scattering cross-section and shape of the target. However, the conventional laser radar uses semiconductor detectors working in the linear avalanche gain mode. Because of its high detection threshold, it needs to receive enough echo photons to obtain the echo signal. Therefore, when the laser radar detects distant objects, it needs a large Power and low repetition frequency lasers provide energy drive, which leads to large problems in traditional lidar in terms of volume, detection distance and detection accuracy of small and far targets. [0003] Photon counting lidar is ...

AI Technical Summary

Problems solved by technology

These algorithms have a certain effect on removing photon noise, but they also have certain defects. For example, the uneven distribution of noise is not considered, and the denoising accuracy is not very high. The selected parameters require prior knowledge of data or rely on past experience, and are universal poor

[0005] To sum up, photon radar has significant advantages in performance compared with lidar, and it will be widely used in various fields around the world. However, photon detection data contains a lot of noise points, and its application is also subject to certain restrictions.

Images