Dense crowd counting method and device

Abstract

The invention provides a dense crowd counting method and a device thereof. By obtaining an image to be detected including a portrait, the image to be detected is input into a convolution neural network model to obtain a crowd density map of the image to be detected, and the number of portraits in the image to be detected is determined according to the crowd density map. The process fully extractsthe feature information from the image to be detected, realizes good crowd counting and density estimation, and brings great convenience for subsequent safety monitoring, crowd control and other applications.

Description

technical field [0001] The invention relates to the technical field of aerial surveillance, in particular to a dense crowd counting method and device. Background technique [0002] With the continuous opening of the country to the use of low-altitude fields in recent years, drones are often used to perform tasks such as auxiliary disaster inspections and special scene surveillance. In these tasks, the statistics and analysis of crowd information is one of the most important links. In many scenes, especially crowded scenes, the number of people and their density distribution in statistical scenes are very important for safety monitoring, emergency warning and intelligence. Many types of tasks such as analysis have an immeasurable role. [0003] In most scenes where crowd counting and density estimation are required, the density of crowds is high, which causes serious occlusion and inconsistent target scales. Early methods generally used detection and regression based method...

AI Technical Summary

Problems solved by technology

However, due to the use of multiple different convolutional neural networks in this type of method, there is a large amount of redundancy in the parameters of the multi-channel branches and the features extracted. The corresponding information cannot be fully extracted, which limits its effect to a certain extent

Images