Fast Human Head and Shoulder Detection Using Convolutional Networks and RGBD Data

Infrared thermography for camera-based skin temperature measurement is increasingly used in medical practice, e.g., to detect fevers and infections, such as recently in the COVID-19 pandemic. This contactless method is a promising technology to continuously monitor the vital signs of patients in clinical environments. In this study, we investigated both skin temperature trend measurement and the extraction of respiration-related chest movements to determine the respiratory rate using low-cost hardware in combination with advanced algorithms. In addition, the frequency of medical examinations or visits to the patients was extracted. We implemented a deep learning-based algorithm for real-time vital sign extraction from thermography images. A clinical trial was conducted to record data from patients on an intensive care unit. The YOLOv4-Tiny object detector was applied to extract image regions containing vital signs (head and chest). The infrared frames were manually labeled for evaluation. Validation was performed on a hold-out test dataset of 6 patients and revealed good detector performance (0.75 intersection over union, 0.94 mean average precision). An optical flow algorithm was used to extract the respiratory rate from the chest region. The results show a mean absolute error of 2.69 bpm. We observed a computational performance of 47 fps on an NVIDIA Jetson Xavier NX module for YOLOv4-Tiny, which proves real-time capability on an embedded GPU system. In conclusion, the proposed method can perform real-time vital sign extraction on a low-cost system-on-module and may thus be a useful method for future contactless vital sign measurements.

Section: Discussionmentioning

confidence: 99%

A Deep Learning-Based Camera Approach for Vital Sign Monitoring Using Thermography Images for ICU Patients

Lyra

Mayer

et al. 2021

“…They used several single-pass and region-based CNN architectures (SDD, R-CNN, Faster R-CNN, and PVANET) with base detectors. El Ahmar et al [22] introduced a real-time approach for the detection of the human head and shoulders from RGB-D data based on image processing and deep learning. They added Candidate Head Locations (CHL) to exploit depth data to improve detection accuracy.…”

Section: Related Workmentioning

confidence: 99%

New End-to-End Strategy Based on DeepLabv3+ Semantic Segmentation for Human Head Detection

Chouai

Doležel

Štursa

et al. 2021

In the field of computer vision, object detection consists of automatically finding objects in images by giving their positions. The most common fields of application are safety systems (pedestrian detection, identification of behavior) and control systems. Another important application is head/person detection, which is the primary material for road safety, rescue, surveillance, etc. In this study, we developed a new approach based on two parallel Deeplapv3+ to improve the performance of the person detection system. For the implementation of our semantic segmentation model, a working methodology with two types of ground truths extracted from the bounding boxes given by the original ground truths was established. The approach has been implemented in our two private datasets as well as in a public dataset. To show the performance of the proposed system, a comparative analysis was carried out on two deep learning semantic segmentation state-of-art models: SegNet and U-Net. By achieving 99.14% of global accuracy, the result demonstrated that the developed strategy could be an efficient way to build a deep neural network model for semantic segmentation. This strategy can be used, not only for the detection of the human head but also be applied in several semantic segmentation applications.

“…In addition, camera-based detection of individuals and the subsequent counting have already confirmed good-to-excellent results in many demonstrated applications, with an accuracy of over 90% [37,38]. In the studies, cameras were either placed above a door or passageway in a top-down view [39] or in a slightly tilted position for better detection of the head and shoulder area [40,41]. In these latter studies, computer vision-based approaches are typically used to recognize people, track them, and thereby determine visitors.…”

Section: Related Workmentioning

confidence: 94%

Testing and Evaluation of Low-Cost Sensors for Developing Open Smart Campus Systems Based on IoT

Neis,

Warch,

Hoppe

2023

Urbanization has led to the need for the intelligent management of various urban challenges, from traffic to energy. In this context, smart campuses and buildings emerge as microcosms of smart cities, offering both opportunities and challenges in technology and communication integration. This study sets itself apart by prioritizing sustainable, adaptable, and reusable solutions through an open-source framework and open data protocols. We utilized the Internet of Things (IoT) and cost-effective sensors to capture real-time data for three different use cases: real-time monitoring of visitor counts, room and parking occupancy, and the collection of environment and climate data. Our analysis revealed that the implementation of the utilized hardware and software combination significantly improved the implementation of open smart campus systems, providing a usable visitor information system for students. Moreover, our focus on data privacy and technological versatility offers valuable insights into real-world applicability and limitations. This study contributes a novel framework that not only drives technological advancements but is also readily adaptable, improvable, and reusable across diverse settings, thereby showcasing the untapped potential of smart, sustainable systems.