Multi-label-based light-weight rapid crowd counting method

Abstract

The invention discloses a multi-label-based lightweight rapid crowd counting method. A simple and efficient trunk feature extraction network is designed according to the size of a receptive field, anda dense context module is arranged in the trunk feature extraction network, so that information transmission of a network layer is ensured, and the network expression capability is improved; six multi-scale intermediate supervision branches are designed, so that the network can be converged more quickly and stably; an up-sampling module is designed, the resolution is improved step by step, and the quality of a density map is improved, so that accurate counting and accurate positioning are realized; three labels are designed, a crowd counting task based on density is explicitly converted intoa foreground and background segmentation task to assist a regression task of a crowd density map, prediction of the density map and the segmentation map is achieved at the same time, and estimation errors are effectively reduced. Test results of UCF _ CC _ 50, ShanghaiTeck and UCF-QNFR data sets show that the prediction performance of the method is superior to that of a current mainstream algorithm, the prediction speed reaches real time, and the method can be conveniently deployed in terminal equipment.

Description

technical field [0001] The invention belongs to the field of crowd counting in computer vision, and is a method for predicting a density map using a convolutional neural network and integrating it to obtain the total number of people in a single picture, which is different from the current mainstream convolution based on VGG, ResNet and DenseNet Network, this method has a simple structure and a small amount of parameters. It is a fast and lightweight crowd counting method based on multi-labels. It has a considerable balance between accuracy and operating efficiency, and has been well verified on public datasets. Background technique [0002] In recent years, large-scale group activities such as protests, festivals, concerts, and sports events have become increasingly frequent, and group emergencies caused by dense crowds have become the focus of society. Crowd counting is an important method of crowd control and management. It can not only count the crowd in the current scen...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the problems that the algorithm complexity in the field of crowd counting is high, it is difficult to achieve real-time, and the accuracy and speed cannot be properly balanced, the present invention designs a light-weight and fast crowd counting neural network based on multi-labels. Efficiency has a considerable balance and is easy to deploy into end devices

Images