An Algorithm of Geomagnetic Detector Based on Neural Network Self-learning

Abstract

The invention discloses a geomagnetic detector detection algorithm based on neural network self-learning. The determination threshold is adaptively updated according to the characteristics of a passing vehicle on the geomagnetism, and thus high accuracy of long-term operation is guaranteed. The determined threshold value Vk is empirically set, the magnetic field change value Ak+1 is recorded after one vehicle passes every time, and the initial threshold is subtracted so as to obtain the threshold change of the bicycle deltaVk+1 and then a certain weight is set for the change amount to correct the initial threshold value Vk. With the increase of the number of vehicle samples, the threshold value is infinitely close to the theoretical determination value and false detection and missed detection can be reduced so that the accuracy of vehicle inspection can be greatly enhanced, the effective and stable traffic flow data can be provided for traffic signal control and the tedious work of periodically and manually adjusting the threshold value of vehicle inspection can be saved.

Description

technical field [0001] The invention relates to the technical field of intelligent transportation, in particular to a detection algorithm for geomagnetic detectors based on neural network self-learning. Background technique [0002] The geomagnetic detector is an information collection method in the intelligent transportation industry. It has the advantages of simple installation, all-weather, high detection accuracy, and good long-term stability. It has been widely used in the collection of traffic flow at urban traffic intersections. The market share is large, and each intersection needs about 10 geomagnetic detectors, and the number of geomagnetic detectors in each city may exceed 2,000. However, the accuracy of geomagnetic detection varies between different manufacturers and vehicle types at different intersections. If each geomagnetic detector is fine-tuned, it will continue to take up a lot of debugging man-hours, and the cost is very high. Some manufacturers cannot co...

AI Technical Summary

Problems solved by technology

However, the accuracy of geomagnetic detection varies between different manufacturers and vehicle types at different intersections. If each geomagnetic detector is fine-tuned, it will continue to take up a lot of debugging man-hours, and the cost is very high. Some manufacturers cannot continue to maintain Geomagnetic detector, but without fine tuning, its accuracy rate cannot be maintained at a high accuracy rate for a long time, which will mislead the adaptive operation of the traffic signal machine, resulting in traffic jams at intersections or low traffic efficiency

Images