Method for forecasting traffic flow in urban area based on depth learning

Abstract

The invention discloses a method for forecasting the traffic flow in the urban area based on depth learning. This method can forecast the traffic flow in each region of the city by extracting the high-dimensional space-time characteristics of traffic flow changes, which provides a new idea for urban traffic flow forecasting. Firstly, according to data of LPR equipment, the historical flow of each time period of the urban area is calculated; then the traffic flow forecasting model is designed by ConvLSTM and CNN, and the flow data of the critical time period which affects the forecast period is extracted as the input training model; and finally, the trained model is used to forecast the traffic flow of the urban area.

Description

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine

Login to view more

AI Technical Summary

Benefits of technology

This patented technique allows us to predict traffic flows over different parts of our country from just about any part of that world without relying solely upon previous data sources like past records. It uses advanced techniques such as convolutional networks and Long Short Term Memory (LSTM) structures to capture complex patterns within large amounts of data. By combining these two types of models together, we aimed towards better understanding how traffic moves across multiple locations during peak hours. Overall, this new approach provides more accurate predictions than current approaches alone.

Problems solved by technology

The technical problem addressed by this patented method relates to improving accurate traffic flow predictors while considering both temporal and spatial aspects during realistic scenarios like highway congested areas.

Images