People/Animal Counting - Integrated Sensor Based and Wifi/Machine Learning Based System

Especially with the COVID-19 pandemic, which is affecting the whole world today, accurate determination of the number of people entering a closed area has become very important in terms of measures to be taken against the spread of the disease. This project uses WiFi Signals and IR sensors to determine the number of people in a predefined area. To obtain the required raw data the system utilizes two main principles: the effect of humans on the WiFi signal strength and the entrance-exit control based on distance and LDR sensors. It is a low cost solution with quite good sensitivity and low error level. The simple principles which it is based on allows its application also in other areas like for example livestock farming, where the number of cattle entering a closed area can be counted. © 2022 IEEE.

Capacity Estimation from Environmental Audio Signals Using Deep Learning

Estimating the capacity of a room or venue is essential to avoid overcrowding that could compromise people’s safety. Having enough free space to guarantee a minimal safety distance between people is also essential for health reasons, as in the current COVID-19 pandemic. Already existing systems for automatic crowd counting are mostly based on image or video data, and some of them, using deep learning architectures. In this paper, we study the viability of already existing Deep Learning Crowd Counting systems and propose new alternatives based on new network architectures containing convolutional layers, exclusively based on the use of environmental audio signals. The proposed architecture is able to infer the actual capacity with a higher accuracy in comparison to previous proposals. Consequently, conclusions from the accuracy obtained with out approach are drawn and the possible scope of deep learning based crowd counting systems is discussed. © 2022, Springer Nature Switzerland AG.

Camera-based Human Counting for COVID-19 Capacity Restriction

In this paper, a Human Counting system is implemented for COVID-19 capacity restrictions. It was implemented using the deep learning model You Only Look Once version 3(YOLOv3) to detect and count the people in a room. The system also can monitor the social distancing between the people in the room while labeling each person as 'safe' or 'unsafe' depending on whether they respect the social distancing protocols that the World Health Organization recommended or not. To make the project user friendly, a Graphical User Interface (GUI) was implemented to allow the user to choose the source of their images that will be used as input to be processed by the system. An experiment was carried out to evaluate the performance of the system under different conditions and in different scenarios where the evaluation was done according to some metrics such as accuracy, precision and recall. The output results from this experiment were demonstrated in details and compared to a similar algorithm as both algorithms focused on people detection using images from an inclined camera. The results show an accuracy of 96% for detection and the number of people counted. © 2022 IEEE.

Real-Time People Counting Using IR-UWB Radar

COVID-19 pandemic has introduced social distance regulations which are crucial to be followed by. In order to maintain proper social distancing, it is critical to regulate the number of people in a closed space. In this paper, we propose a people counting system based on Impulse Radio Ultra-Wideband radars for counting people walking through a doorway. The system uses two IR-UWB radars placed horizontally apart to create a lag effect when someone walks by the radars. This enables detection of movement’s direction and subsequently, determination of the number of people in a room. The system proposed can be used for people counting in real-time and also on saved data which offers flexibility for real world applications. Several tests were conducted which shows the accuracy rate of system to be around 90%, validating the system. Contrary to conventional vision based people counting system, the proposed system is not limited by environmental factors such as light and also is privacy oriented. © 2022, ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering.