Method, device and equipment for counting number of dense people and storage medium

Abstract

The invention discloses a method, device and equipment for counting number of dense people and a computer readable storage medium, and the method comprises the steps: inputting a to-be-tested image into a pre-trained target residual multi-scale convolutional neural network, and enabling a residual unit to be arranged between two pre-selected convolutional layers in the target residual multi-scaleconvolutional neural network; processing the to-be-tested image by using the target residual multi-scale convolutional neural network to obtain an estimated density map of the to-be-tested image, andoutputting the estimated density map; and determining the number of people in the to-be-tested image according to the target estimation density map. According to the method, the device, the equipmentand the computer readable storage medium provided by the invention, the degradation problem of the network is effectively solved by utilizing the residual error unit, so that the counting error of thesecret-level crowd is effectively reduced.

Description

Technical field [0001] The present invention relates to the field of image processing technology, and in particular to a method, device, equipment and computer-readable storage medium for counting the number of people in secret. Background technique [0002] The task of crowd counting in computer vision is to automatically estimate the number of people in an image or video. In order to help control crowd numbers and public safety in many scenarios such as public gatherings and sports events, accurate crowd counting becomes increasingly important. The traditional dense crowd counting methods include two types: detection-based methods and regression-based methods. However, detection-based methods will be affected by crowd occlusion, while regression-based methods are often affected by sharp changes in viewing angle and scale. In view of the shortcomings of the above two traditional methods, the method based on deep learning convolutional neural network is used in the estimation o...

AI Technical Summary

Problems solved by technology

[0005] The object of the present invention is to provide a method, device, equipment and computer-readable storage medium for counting dense crowds, so as to solve the problem that the multi-layer convolutional neural network in the prior art will cause degradation problems during training

Images