Block pedestrian flow prediction method and system based on space-time diagram convolutional neural network

Abstract

The embodiment of the invention provides a block pedestrian flow prediction method and system based on a space-time diagram convolutional neural network. The method comprises the following steps: acquiring historical movement track data of pedestrian flow in a block; inputting the block pedestrian flow movement track data into a pre-trained block pedestrian flow prediction model to obtain a blockfuture pedestrian flow prediction result output by the block pedestrian flow prediction model, wherein the block pedestrian flow prediction model is obtained by constructing a dynamic space-time diagram, classifying blocks by using interest points and training a three-dimensional diagram convolutional neural network model. According to the embodiment of the invention, the method and system achievethe better pedestrian flow prediction performance through the construction of the dynamic graph representing the spatial correlation and time dynamics, the representation of the functional attributesof different blocks through the interest points, and the simultaneous prediction of pedestrian flow and block functions through the multi-task learning.

Description

technical field [0001] The invention relates to the technical field of flow forecasting, in particular to a method and system for predicting pedestrian flow in a block based on a spatio-temporal graph convolutional neural network. Background technique [0002] According to statistics from the Population Division of the United Nations Department of Economic and Social Affairs, as of 2018, 55% of the world's population lived in cities, and with further urbanization, this figure will increase by 68% by 2050. With the rapid growth of urban population, more and more problems have begun to emerge: urban traffic congestion, urban infrastructure overwhelmed, and crowd stampede accidents are becoming more frequent. In order to solve a series of problems associated with urbanization, urban people (vehicle) flow prediction has received extensive attention from researchers in recent years. Among them, urban people flow prediction mainly studies the population flow patterns between diffe...

AI Technical Summary

Problems solved by technology

[0007] The embodiment of the present invention provides a block crowd flow prediction method and system based on spatio-temporal graph convolutional neural network, which is used to solve the problem that the solutions for crowd flow prediction in the prior art cannot adapt to complex dynamic patterns, and the accuracy rate is low, so it is not suitable for Irregular area scenarios cannot take into account the influence factors of time and space at the same time, and do not meet the actual needs of online operation

Images