Rail transit passenger flow abnormal mode identification method based on big data

Abstract

The invention discloses a rail transit passenger flow abnormal mode identification method based on big data. The rail transit passenger flow abnormal mode identification method comprises the steps of:acquiring card swiping information of a passenger, and judging whether a card belongs to an abnormal card or not, wherein the abnormal card is used for representing that abnormal behaviors exist in card swiping records; and if the card is judged to be an abnormal card, acquiring face image information and human body characteristics of the abnormal card passenger; binding the abnormal card numberwith the face image information; and publishing the face image information corresponding to the abnormal card to a plurality of tracking cameras, respectively comparing the face image information acquired by the plurality of tracking cameras according to a local face recognition library by means of the plurality of tracking cameras, and if the face information in the local face recognition library is retrieved, uploading a comparison result and a video stream of the corresponding camera to a target server. According to the rail transit passenger flow abnormal mode identification method, abnormal cards and abnormal behaviors can be accurately identified, and abnormal behavior paths and time periods can be counted and analyzed.

Description

technical field [0001] The present invention relates to the technical field of rail transit passenger flow identification, and more particularly, to a method for identifying abnormal patterns of rail transit passenger flow based on big data. Background technique [0002] In recent years, with the popularization of transportation cards and the development of computer science and technology, a large number of pedestrian travel data in cities have been collected. In this case, the research on pedestrian travel behavior has entered the era of big data. By analyzing the card data, we can further understand the rules of pedestrian travel behavior, so as to provide decision-making assistance for urban safety management departments. [0003] At present, there are mainly two ideas in the process of applying deep learning to the target tracking process of pedestrian travel behavior: first, using the transferability of the features learned by the deep neural network, first in large-sc...

AI Technical Summary

Problems solved by technology

[0005] 1) Complex background interference: The environment of the tracked target will affect the accuracy of tracking. For example, the background is similar in color to the target, or objects similar to the target appear, which may lead to tracking errors

[0006] 2) Changes in the appearance of the target: During video tracking, the brightness of the light and the shape of the target may change, and the sudden loss of focus during shooting will cause the edge of the target to be blurred

The tracking of local information is easy to lead to inaccuracy, and when the target is fully occluded, it is also difficult to quickly find the previous target again

Images