A road abnormity detection model based on window partition and dynamic time regularization

Abstract

The invention relates to a road abnormality detection model based on window partition and dynamic time regularization, comprising the following steps: 1) carrying out threshold detection and sliding window processingon z-axis acceleration data, and screening fragments to be determined; 2) comparing the acceleration data of the segment to be determined with the segments of several known abnormal types and normal road segments through a dynamic time regularization algorithm to obtain a difference degree vector; 3) determining the abnormal type of the segment to be determined. The road abnormality detection model has the beneficial effects that the abnormal section of the road can be intercepted completely, and the experiment result also proves that the abnormal condition of the road can be detected more accurately on different data sets, and the effect of the method is better than that of the existing method in terms of the two-classification or the multi-classification.

Description

technical field [0001] The invention relates to a road detection model, in particular to a road detection model based on window division and dynamic time warping. Background technique [0002] Highways play an important backbone and pillar role in my country's comprehensive transportation network. By the end of 2016, the total mileage of national highways reached 4.6963 million kilometers. It is estimated that my country's road network will have a scale of about 5.8 million kilometers by 2030. With the rapid development of road construction, the demand for road maintenance and management is also growing rapidly. Governments of various countries will spend huge manpower and material resources on road maintenance. For example, the British government announced that it spent 1.2 billion US dollars on road maintenance in 2017. The City of Toronto spent a total of $6 million on repairing potholes. Road damage can lead to serious traffic accidents, for example, between 2000 and 20...

AI Technical Summary

Problems solved by technology

First of all, some of the current methods are mostly based on sliding windows to process the entire section of data, and road anomalies are often unevenly distributed in time series data, and at the same time their lengths are also different, so to a large extent there is a large number of road anomalies. Possibility of split

Secondly, some current methods rely on data sets and features, and often perform very differently on different data sets. Multi-classification methods are also difficult to apply to binary classification data sets.

Images