A scene monitoring method based on two-dimensional laser range finder array

Abstract

The invention relates to a scene monitoring method based on a two-dimensional laser rangefinder array. The method uses multiple lasers to realize scene monitoring, overcomes the influence of weather and other environments, has a wide monitoring range, and fast operation speed. It mainly includes multi-laser data fusion and target Detection (Figure 4 is a summary drawing). In multi-laser joint calibration, a multi-laser joint calibration method based on a virtual dihedron is proposed. Compared with taking points for calibration, it has high accuracy, simple operation and strong practicability. The thickness is required, so this method is especially suitable for practical applications. In terms of target extraction, single-Gaussian background modeling is improved and adapted to laser data. The measurement error of the laser range finder is related to the size of the distance value. Compared with the fixed threshold when judging whether it is the background or the foreground, the single Gaussian background modeling is adaptive. Compared with the histogram method, it can overcome the influence caused by some stationary pedestrians.

Description

technical field [0001] The invention relates to the technical field of scene monitoring, in particular to a scene monitoring method based on a two-dimensional laser range finder array. Background technique [0002] Analyzing or monitoring human activities, such as counting the number of passengers or measuring trajectories, etc., is widely used in various fields, such as building security, shopping mall management, railway station and so on. Analysis of video data has been the main method of surveillance until now. In video surveillance systems, cameras are placed at specific locations to reduce occlusion. With such a camera setup, the resolution and viewing angle of the image are rather limited, and moving objects with fewer image pixels may be lost in tracking. Also, constant changes in lighting and weather conditions degrade the reliability and robustness of the vision system. Achieving all-round monitoring requires the use of multiple cameras. For example, if the fie...

AI Technical Summary

Problems solved by technology

In the process of completing multi-laser data fusion, CN101387700 discloses a characteristic that the direction vector of the laser point measured by a laser scanner placed in any position and direction in the same plane matches the plane model of the moving target. Laser scanning data fusion, this method needs to pre-define a plane profile model of the moving target, when the moving target diversification will increase the complexity of the algorithm

CN102253391A discloses a pedestrian target tracking method based on multiple laser radars, which uses a multi-point method to achieve calibration between multiple lasers, but it is difficult to obtain points for the experiment, and if the calibration rod is too thick, the corresponding laser points will be inaccurate , increasing the calibration error, if it is too thin, it is easy to cause missing scans

There is a certain interval between the scanning beams of the laser rangefinder, and as the distance increases, it diverges, which will also lead to missing scans

Using the histogram method to obtain background data, the most frequent occurrence is background point data, and it is easy to determine pedestrians who have been stationary for a long time as background data, which reduces the accuracy

To sum up, there is no mature method to achieve multi-laser joint calibration and target detection

Images