Fully Optimized Convolutional Neural Network Based on Small-Scale Crowd

Crowd counting is of considerable significance to society in terms of public safety and urban development. Manual counting of people in a video or photo is often time-consuming and labour-intensive. People will need an efficient and economy way instead of counting manually. Nowadays, the convolutional neural network was popularly utilized as the baseline for crowd counting. However, the more complex the CNN-based algorithm, the more computing resources will be consumed. This article aims to present a simpler and faster fully optimized convolutional neural network for crowd counting with desired performance. To minimize the computational cost on training networks, we proposed a fully optimized method to build our network. Extensive experiments on our fully optimized convolutional neural network indicate the superiority of our network that has very high accuracy and speed on small scale crowd.

I. Introduction

Crowd counting is to count the number of people in a specific area. It is of great value in public, and crowd counting is also an important research topic in the field of computer vision and intelligent video surveillance. It aims to build a high-performance cognitive system for crowd monitoring and scene understanding [1]-[3]. First, it has a huge benefit of social management and safety. For example, monitoring the flow of people in a square or a mall in a festival to prevent accidents[4], [5]. Meanwhile, the research on the crowd also has a certain economic effect. For example, investigating the crowd of people in a mall could be used to determine their business capabilities. Moreover, the study of crowd counting also affects other different subjects like psychology [6] and biology [7], [8].