Efficient Centroid Based Distance Monitoring System Using Deep Learning

COVID-19 virus spread very rapidly if we come in contact to the other person who is infected, this was treated as acute pandemic. As per the data available at WHO more than 663 million infected cases reported and 6.7 million deaths are confirmed worldwide till Dec, 2022. On the basis of this big reported number, we can say that ignorance can cause harm to the people worldwide. Most of the people are vaccinated now but as per standard guideline of WHO social distancing is best practiced to avoid spreading of COVID-19 variants. This is difficult to monitor manually by analyzing the persons live cameras feed. Therefore, there is a need to develop an automated Artificial Intelligence based System that detects and track humans for monitoring. To accomplish this task, many deep learning models have been proposed to calculate distance among each pair of human objects detected in each frame. This paper presents an efficient deep learning monitoring system by considering distance as well as velocity of the object detected to avoid each frame processing to improve the computation complexity in term of frames/second. The detected human object closer to some allowed limit (1m) marked by red color and all other object marked with green color. The comparison of with and without direction consideration is presented and average efficiency found 20.08 FPS (frame/Second) and 22.98 FPS respectively, which is 14.44% faster as well as preserve the accuracy of detection. © 2023 IEEE.

A Smart Social Distance Monitoring System

The entire world is affected by Covid-19 pandemic. One of the major reasons is that it is contagious and a minimum distance should be maintained to stay safe. Social distancing might be a difficult task to implement effectively. Social distancing plays a pivotal role in curbing diseases that are contagious like Covid-19.Now that situations are returning to normal, the risk of getting infected is still high. Governments are deciding to ease lockdown regulations, as part of the unlocking public places, workspaces and educational institutions started to resume their activities. Considering the current scenario, the public has to strictly follow all the necessary Covid-19 protocols to reduce the spike in the number of Covid cases. This project aims to develop a prototype device that helps in implementing social distancing using Ultra-Wide Band (UWB) wireless technology based solution. Prototype issues an alert signal when the distance between individuals is less than the prescribed threshold distance. If the protocol is breached, the user is alarmed through an LED. UWB is known for its advantages as it has greater signal strength compared to Bluetooth. The design of the prototype enables implementation as wearable such as an ID card. © 2022 IEEE.

YOLO KeepSafe: Utilization of YOLOv4-Based Algorithm to Employ a Real-time People Detection and Social Distance Monitoring System

This paper outlines the process used to create the social distance detection system - YOLO KeepSafe. The researchers discuss several calibration procedures, investigate the difficulties that arise when used in real-time situations, and offer potential solutions. © 2022 IEEE.

Social Distance Monitoring Framework Using YOLO V5 Deep Architecture

Due to the current outburst and speedy spread of the COVID-19 pandemic, there is a need to comply with social distancing rules by the general public. The public needs to, at minimum, hold a distance of 3 ft or 1 m among one another to follow strict social distancing as instructed by using the World Health Organization for general public safety. Researchers have proposed many solutions based on deep learning to reduce the current pandemic, including COVID-19 screening, diagnosis, social distancing monitoring, etc. This work focuses explicitly on social distancing monitoring by a deep learning approach. Here we employ the YOLOV5 object detection technique upon different images and videos to develop a strategy to assist and put strict social distancing in public. The YOLOV5 algorithm is more robust and has a quicker detection pace than its competitors. The suggested object detection framework shows an average precision rating of 94.75%. Some of the existing analyses suffer to identify humans within a range. A few identification blunders happen because of overlapping video frames or humans taking walks too near each other. This detection mistake is due to the overlapping structures, and human beings are standing too close to each other. This paper focuses on correctly identifying humans by using and overcoming the flaws of frame overlapping. Following the proposed social distancing technique also yields positive results in numerous variable eventualities. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

A Hybrid Deep Learning System to Detect Face-Mask and Monitor Social Distance

Coronavirus Disease 2019 (COVID-19) emerged towards the end of 2019, and it is still causing havoc on the lives and businesses of millions of people in 2022. As the globe recovers from the epidemic and intends to return to normalcy, there is a spike of anxiety among those who expect to resume their everyday routines in person.The biggest difficulty is that no effective therapeutics have yet been reported. According to the World Health Organization (WHO), wearing a face mask and keeping a social distance of at least 2 m can limit viral transmission from person to person. In this paper, a deep learning-based hybrid system for face mask identification and social distance monitoring is developed. In the OpenCV environment, MobileNetV2 is utilized to identify face masks, while YoLoV3 is used for social distance monitoring. The proposed system achieved an accuracy of 0.99. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

A Social Distance Monitoring Method Based on Improved YOLOv4 for Surveillance Videos

Social distance monitoring is of great significance for public health in the era of COVID-19 pandemic. However, existing monitoring methods cannot effectively detect social distance in terms of efficiency, accuracy, and robustness. In this paper, we proposed a social distance monitoring method based on an improved YOLOv4 algorithm. Specifically, our method constructs and pre-processes a dataset. Afterwards, our method screens the valid samples and improves the K-means clustering algorithm based on the IoU distance. Then, our method detects the target pedestrians using a trained improved YOLOv4 algorithm and gets the pedestrian target detection frame location information. Finally, our method defines the observation depth parameters, generates the 3D feature space, and clusters the offending aggregation groups based on the L2 parametric distance to finally realize the pedestrian social distance monitoring of 2D video. Experiments show that the proposed social distance monitoring method based on improved YOLOv4 can accurately detect pedestrian target locations in video images, where the pre-processing operation and improved K-means algorithm can improve the pedestrian target detection accuracy. Our method can cluster the offending groups without going through calibration mapping transformation to realize the pedestrian social distance monitoring of 2D videos. © 2023 World Scientific Publishing Company.

Social Distance Monitoring Framework Using YOLO V5 Deep Architecture

Due to the current outburst and speedy spread of the COVID-19 pandemic, there is a need to comply with social distancing rules by the general public. The public needs to, at minimum, hold a distance of 3 ft or 1 m among one another to follow strict social distancing as instructed by using the World Health Organization for general public safety. Researchers have proposed many solutions based on deep learning to reduce the current pandemic, including COVID-19 screening, diagnosis, social distancing monitoring, etc. This work focuses explicitly on social distancing monitoring by a deep learning approach. Here we employ the YOLOV5 object detection technique upon different images and videos to develop a strategy to assist and put strict social distancing in public. The YOLOV5 algorithm is more robust and has a quicker detection pace than its competitors. The suggested object detection framework shows an average precision rating of 94.75%. Some of the existing analyses suffer to identify humans within a range. A few identification blunders happen because of overlapping video frames or humans taking walks too near each other. This detection mistake is due to the overlapping structures, and human beings are standing too close to each other. This paper focuses on correctly identifying humans by using and overcoming the flaws of frame overlapping. Following the proposed social distancing technique also yields positive results in numerous variable eventualities. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Social Distance Monitoring System Using YOLO and Pixel-to-real-world Distance Mapping

Covid-19 is an infectious disease caused by the recently discovered coronavirus. This virus spreads through droplets produced when an infected person coughs, sneezes, or exhales. A person can be infected by breathing air containing the virus if a person is too close to someone who is already infected with Covid-19. A person can also be infected by touching contaminated surfaces and then touching their eyes, nose, or mouth. Social distancing is to reduce the spread of viruses, other than washing hands and wearing a mask. With the help of Computer Vision Technology, it can monitor the safe distance of human activities in a particular area or environment. The problem is to determine the safe distance in a pixel-based digital image. The difference in the same pixel distance does not always mean the actual distance between adjacent objects is the same. It must consider the actual distance to the camera. Since social distancing requires someone to keep a certain distance from another, a web-based application with a Convolutional Neural Network (CNN) algorithm is employed using the You Only Look Once (YOLO) and Pixel-to-real-world distance mapping technique. In testing, there are several test scenarios with the accuracy results obtained of 95% with a recall of 0.95 and a precision of 0.92, and MAE 5.9 cm. © 2022 IEEE.

Edge-Oriented Social Distance Monitoring System Based on MTCNN

Social distance monitoring (SDM) systems are vital in fighting the spread of the coronavirus (COVID-19). Existing SDM systems employ bounding box-method, which imposes inaccurate distance estimation due to the high variance in its output coordinates. To solve this problem, an SDM system based on multitask cascaded convolutional neural networks (MTCNN) is proposed. Instead of using bounding box coordinates, face detection and facial landmarks localization of MTCNN is used to provide fixed coordinates and increase the distance estimation accuracy of SDM. However, while the accuracy issue is solved by using MTCNN, the SDM system suffer from large computational requirements due to the cascaded networks added on top of the distance estimation process. To deal with this challenge, a constrained optimization technique is employed to each stage of MTCNN with the goal of reducing its hardware requirements while keeping the same reliability as the original implementation. Experimental results show that the SDM system based on the optimized MTCNN achieves higher accuracy performance with reduced computational requirements as compared to conventional SDM systems. This allows the proposed SDM system using optimized MTCNN to be deployed efficiently on edge devices. IEEE

Social Distance Monitoring to Enhance the Safety from COVID-19 Infection

The best precaution in the COVID-19 pandemic period is social distancing. Despite being aware of these, people often violate social distance and take down their masks in public places, knowingly or unknowingly. This unawareness is the primary reason for public places like retail shops, bus stops, public transport, educational institutions, hospitals, and clinics to become increasingly inaccessible and COVID-19 hotspots. It would be practically impossible for anyone to make sure everyone is following social distancing in a public place;a human will not be able to see beyond his field of view. An automated CCTV-based real-time system could help in monitoring the crowd. The proposed model takes a video stream as input, segments people using a CNN-object detection algorithm (YOLO), anchors objects' locations, and finds the distance between them. The system then marks bounding boxes around people who do not follow social distancing. Using pixel coordinates of violating people can accurately mark and alert security personnel or display wherever social distancing is violated. © 2022 IEEE.

IoT and AI-enabled Physical Distance Monitoring Application to Prevent COVID19 Transmission

During COVID19 pandemic, people are encouraged to practice physical distancing at least 1 meter when interacting with other people to prevent the spread of the COVID19. This study aims to develop a system that can monitor the physical distancing and track physical contact in a room using internet of things (IoT) and artificial intelligent technology. The system consists of a small single-board computer (Raspberry Pi), webcam, and web application displaying physical contact information. The system uses YOLO algorithms to detect the human object and euclidean distance formula to determine the distance between human objects. We evaluated the performance of YOLOv3 and YOLOv3-tiny running on Raspberry Pi. The evaluation result shows that YOLOv3 consumes more CPU resources than YOLOv3-tiny but has better accuracy in detecting human objects. YOLOv3-tiny can process images and detect objects faster than YOLOv3. © 2022 IEEE.

Camera-based distance detection and contact tracing to monitor potential spread of COVID-19

Purpose: This study aims to analyze a social distance monitoring and contact tracing assistance tool for preventing the spread of COVID-19 in a busy indoor working hospital environment. Method: A camera-based tool was developed. The tool estimates physical distance between multiple individuals in real-time and also tracks individuals and records their contact time when in violation social distance requirements for retrospect review. Both stereo- and monocular-camera tools are implemented and their accuracy and efficiency are evaluated and compared. Video was captured by a ZED M camera which was set close to the ceiling of a lab space. Three people within the field of view of the camera completed various movements. The distance (binary, <6 feet or >6 feet) and contact time between each pair was recorded as ground truth and compared to the video software analysis. Additionally, the contact time between any two individuals was calculated and compared to ground truth. Results: The overall accuracy of social distance detection was 95.1% and 74.4%, with a false-negative rate (when the tool predicts individuals are far enough apart, when they are actually too close) of 7.2% and 23.5% for the stereo and monocular tools, respectively. Conclusions: A stereo-camera social distance monitoring and contact tracing assistance tool can accurately detect social distance among multiple people, and keep an accurate contact record for each individual. While a monocular camera tool provided some level of certainty, a stereo camera tool was shown to be superior.

Artificial Intelligence Based Social Distance Monitoring in Public Areas

COVID-19 is an infectious disease caused by a newly discovered coronavirus called SARSCoV-2. There are two ways of contamination risk, namely spreading through droplets or aerosol-type spreading into the air with people's speech in crowded environments. The best way to prevent the spread of COVID-19 in a crowd public area is to follow social distance rules. Violation of the social distance is a common situation in areas where people frequently visit such as hospitals, schools and shopping centers. In this study, an artificial intelligence-based social distance determination study was developed in order to detect social distance violations in crowded areas. Within the scope of the study, a new dataset was proposed to determine social distance between pedestrians. The YOLOv3 algorithm, which is very successful in object detection, was compared with the SSD-MobileNET, which is considered to be a light weighted model, and the traditionally handcrafted methods Haar-like cascade and HOG methods. Inability to obtain depth information, which is one of the biggest problems encountered in monocular cameras, has been tried to be eliminated by perspective transformation. In this way, the social distance violation detected in specific area is notified by the system to the relevant people with a warning. © 2022 Lavoisier. All rights reserved.

Energy-efficient and Privacy-aware Social Distance Monitoring with Low-resolution Infrared Sensors and Adaptive Inference

Low-resolution infrared (IR) Sensors combined with machine learning (ML) can be leveraged to implement privacy-preserving social distance monitoring solutions in indoor spaces. However, the need of executing these applications on Internet of Things (IoT) edge nodes makes energy consumption critical. In this work, we propose an energy-efficient adaptive inference solution consisting of the cascade of a simple wake-up trigger and a 8-bit quantized Convolutional Neural Network (CNN), which is only invoked for difficult-to-classify frames. Deploying such adaptive system on a IoT Microcontroller, we show that, when processing the output of a 8 × 8 low-resolution IR sensor, we are able to reduce the energy consumption by 37-57% with respect to a static CNN-based approach, with an accuracy drop of less than 2% (83% balanced accuracy). © 2022 IEEE.

Social Distance Monitoring and Face Mask Detection Using Deep Learning

As there is no evidence of COVID-19 slowing down in several components of the world, maintaining “social distancing” (also mentioned as physical distance) between indoor and outside people is more vital than ever. It is counseled that two people keep a distance of 1.8 m (approximately, vi feet) apart. Python can be used to capture people and monitor social distancing. Deep learning, TensorFlow, Keras, and OpenCV are used to acknowledge masks;it uses a good computer vision-based technique that focuses on the automatic period observance of individuals to sight safe social distance twenty-four hours a day, seven days every week, for inspectors in public places, commercial centers, and different locations. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Social Distance Monitoring Method with Deep Learning to prevent Contamination Spread of Coronavirus Disease

The ongoing COVID-19 virus pandemic has resulted in a global tragedy due to its lethal spread. The population's vulnerability grows as a result of a lack of effective helping agents and vaccines against the virus. The spread of viruses can be mitigated by minimizing close connections between people. Social distancing is a critical containment tool for COVID-19 prevention. In this paper, the social distancing violations that are being made by the people when they are in public places are detected. As per CDC (Centers for Disease Control and Prevention) minimum distance that should be maintained by people is 2-3 meters to prevent the spread of COVID- 19, the proposed tool will be used to detect the people who are maintaining less than 2-3 meters of distance between themselves and record them as a violation. As a result, the goal of this work is to develop a deep learning-based system for object detection and tracking models in social distancing detection. For object detection models, You Only Look Once, Version 3 (YOLO v3) is used in conjunction with deep sort algorithms to balance speed and accuracy. To recognize persons in video segments, the approach applies the YOLOv3 object recognition paradigm. An efficient computer vision-based approach centered on legitimate continuous tracking of individuals is presented to determine supportive social distancing in public locations by creating a model to generate a supportive climate that contributes to public safety and detect violations through camera. © 2022 IEEE.

Design of an IoT Safety Distance Monitoring Device for COVID-19

The COVID-19 pandemic arises in the year 2020 that leads to almost more than 100,000,000 confirmed cases happened in worldwide and yet, the cases are continuing to rise now. One of the common methods that used to prevent the COVID-19 is through keeping the social distancing between each other. However, people do not aware of the importance of applying social distancing rules. A low-cost solution is urgently needed for contact tracing and remind the users to keep the social distancing with the others. An IoT-based safety distance monitoring (ISDM) device is proposed to monitor safety distance, remind people to always stay alert and keep the data of the users that violated the social distancing rules for contact tracing. ISDM is a lightweight wearable wristwatch. Bluetooth Low Energy technique is used to send and receive information such as body temperature and contact tracing information. Fever is the most common symptom of COVID-19 patients. ISDM can operate either online or offline. The health and contact tracing information are stored in a local Micro-SD card storage when offline and data will be uploaded to the cloud server for further analysis once ISDM is connected. The detailed contact tracing information can be used to trace the potential COVID-19 patients with close contact with the COVID-19 patient. © 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

A Deep learning based approach for Social Distance Monitoring

The COVID-19 pandemic has hit the world at large claiming large number of lives till date leaving us with no solution except maintaining social distancing or washing hands regularly, wearing masks and staying at homes. Social distancing is one of the key aspects to prevent spreading of this virus. It means more of maintaining suitable distance between each other. Artificial intelligence has been used widely for a large number of purposes and as such is one of the key tools used here for implementing this project. The proposed system identifies people who are not suitable distance apart by using object detection and calculating the Euclidian distance between two people. This system would be beneficial to the authorities for alerting people if the situation is serious. © 2021 IEEE.

A Computer Vision Toolkit for Adapting to a COVID-19 Lifestyle

With the fomenting situation that the COVID-19 pandemic brought with it since 2020, the need to adapt to a cautious lifestyle is still the need of the hour. According to the CDC, washing hands frequently, maintaining social distancing and wearing facial coverings are essential precautions one must follow. But with everything that’s going on, one can easily forget to follow one. This brings an opportunity for utilizing computer vision aid to adapt to a “COVID lifestyle”. Taking this motivation into consideration, we developed a computer vision system/toolkit that supports facial-mask detection, social distance monitoring and setting hand washing reminders. A novel facial feature-based data creation was proposed for training and testing our CNN model. The modules in our toolkit are highly portable, making it easy for users to incorporate them in their programs/utilities. We also ensure privacy and security while transferring the camera data stream in a server-client network. © 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.

Human Localization Using a Single Camera Towards Social Distance Monitoring During Sports

Coronavirus disease 2019 (COVID-19) is still prevalent in the world. Social distancing is more important during exercise because we may not be able to wear masks to avoid breathing problems, heatstroke, etc. For supporting management of social distancing, we are developing a human localization system using a single camera especially for sports schools and gyms. We rely on a single camera because of the deployment cost. The system recognizes people from a video and estimates the human positions for supporting management of social distancing. The challenge is the error owing to pose variation during sports. In order to solve the problem, we adjust the height of the waist according to the pose of the legs. For evaluation, we collected 80 images with 5 kinds of poses. The results show that we successfully reduce the absolute position error by 23 cm on average. © 2022, ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering.