A group traffic forecasting model and a method based on a focusing mechanism

Abstract

The invention discloses a group flow forecasting model and a method based on a focusing mechanism, The model comprises a continuity feature learning module, which is used for learning the continuity feature expression by using an Attentive Crowd Flow Machines (ACFM) to the time-sorted feature graph sequence to obtain the continuity feature graph; The periodic feature learning module is used for learning the periodic feature expression of the time-sorted feature graph sequence by using the focused group flow machine ACFM to obtain the periodic feature graph; the periodic feature learning moduleis used for learning the periodic feature expression of the time-sorted feature graph sequence to obtain the periodic feature graph. A fusion module with time variation is used for introducing external information to guide the fusion of the continuous feature map and the periodic feature map. The present invention deduces the future trend of the group traffic by learning the dynamic representation of the change of the data in the time domain.

Description

technical field [0001] The invention relates to the fields of security monitoring, urban traffic management, computer vision, etc., and in particular to a group flow prediction model and method based on a focusing mechanism. Background technique [0002] Group traffic analysis is an important task. Due to its great potential in many intelligent applications, it has attracted a lot of research interest. Such as Figure 1a As shown, in urban management, the goal of city-wide group flow analysis is to predict the future inflow and outflow of urban areas based on GPS signals in the current time period, while in security surveillance, such as Figure 1b As shown, the goal of crowd flow analysis is to predict their number and location by using their current location in the video. Although the regional scales of group flow analysis in different fields vary widely, their core issues all lie in how to mine salient spatial information and construct temporal correlations for more acc...

AI Technical Summary

Problems solved by technology

[0003] In academic research, scholars have proposed a lot of work on space-time modeling, but there are still some challenges that limit the performance of crowd flow analysis in complex scenes

First, in the time series, the group flow data will change greatly, and it is very important to capture this dynamic change; second, some periodic regularity (for example, the traffic flow changes suddenly due to the impact of peak hours or before holidays ) will greatly affect the variation of group flow, which increases the difficulty of learning the characteristics of group flow from data

Images