Crowd density estimation method based on deep learning

Abstract

The invention discloses a crowd density estimation method based on deep learning. The crowd density estimation method comprises the following steps: S1, labeling a training set; S2, generating a firstcrowd density map by utilizing the labeled information; S3, constructing a deep neural network model, and extracting features of different layers through the deep neural network model to establish amulti-layer feature pyramid; analyzing the output features of the feature pyramid to obtain a second crowd density map; S4, training the constructed neural network model by using the first crowd density map generated in the S2 to obtain a trained deep neural network model; and S5, inputting a to-be-estimated image into the trained deep neural network model to obtain a final crowd density map of the to-be-estimated image, and integrating the final crowd density map to obtain the estimated number of people of the to-be-estimated image. The crowd density estimation method based on deep learning estimates crowd density by fusing pyramid features of different layers, and has the advantages of being high in robustness and good in performance.

Description

technical field [0001] The present invention relates to the technical field of computer vision, in particular to a method for estimating airport crowd density based on deep learning. Background technique [0002] With the convenience of transportation, there are more and more crowded places such as railway stations and airports, and the safety hazards are prominent. Estimating the number of people in videos or images through technologies such as computer vision can help evacuate overcrowded crowds, prevent stampede incidents, and give early warnings. [0003] Current people counting methods can be summarized into three categories: object detection-based methods, regression-based methods and deep learning-based methods. The method based on target detection first detects each individual in the crowd, and then counts them to obtain the crowd density. This method is difficult to cope with medium and high density scenes, and the time required for detection is relatively long. ...

AI Technical Summary

Problems solved by technology

However, the current deep learning algorithm generally uses the method of marking the coordinates of the center of the head, which loses a lot of position information; the generated density map does not contain the information of the size of the head; in addition, multi-column convolutional neural network structure is often used, which has high complexity. The problem of large sample requirement and long training time

Images