Related references
Note: Only part of the references are listed.
Article
Computer Science, Information Systems
Shushan Hu et al.
Summary: This article develops a novel deep-learning-based approach for better building occupancy detection based on CCTV cameras. The approach consists of two main modules, feature extraction and three-stage occupancy detection, and achieves superior performance compared to baseline models according to empirical experiments.
IEEE INTERNET OF THINGS JOURNAL
(2023)
Article
Automation & Control Systems
Aya Nabil Sayed et al.
Summary: This paper presents an innovative non-intrusive occupancy detection approach using environmental sensor data, which can aid in energy preservation while maintaining end-user comfort level.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2023)
Article
Construction & Building Technology
Yingdong He et al.
Summary: This study proposes a method to predict occupants' thermal state by utilizing infrared thermography, computer vision, and machine learning. By measuring the distribution of skin temperatures on specific areas of the face and hands, and using temperature differences within and between these areas, the effects of calibration drift in thermal infrared cameras are eliminated. The results show that these measurements accurately predict occupant thermal state.
BUILDING AND ENVIRONMENT
(2023)
Article
Thermodynamics
Yejin Hong et al.
Summary: This study proposes a novel symbolic hierarchical clustering method (HOS-SAX) to evaluate the efficiency and energy usage patterns of building systems. The analysis shows that approximately 71% of the entire operation period is characterized by inefficient operation, and a supply temperature reduction of 0.87 degrees C is expected in the most inefficient sections.
Article
Construction & Building Technology
Yuren Yang et al.
Summary: Providing accurate occupancy schedules for building energy simulation is of great energy-saving significance. However, the standard reference schedules provided by documents are fixed and differ significantly from the actual occupancy fluctuations. This paper proposes a framework using Convolutional Neural Network-based density estimation methods to provide accurate and flexible occupancy schedules. A case study on educational buildings demonstrates the effectiveness of the framework in improving the accuracy of building energy simulation.
BUILDING AND ENVIRONMENT
(2022)
Article
Construction & Building Technology
Samr Ali et al.
Summary: Occupancy detection and estimation are key research areas in smart buildings, with recent advancements in machine learning approaches. This paper introduces a novel method for the use of HMMs in occupancy applications, promising scalable and stable deployment with significant improvements in evaluation metrics.
ENERGY AND BUILDINGS
(2022)
Article
Computer Science, Artificial Intelligence
Joy Dutta et al.
Summary: The OccupancySense model utilizes a unique approach by fusing indoor air quality data, static and dynamic context data to detect human presence and predict indoor occupancy count. With the integration of CatBoost algorithm, the model achieves higher forecasting accuracy compared to other commonly used machine learning algorithms. This non-intrusive model demonstrates high predictive power in accurately detecting occupancy and predicting headcount and occupancy density of the room.
APPLIED SOFT COMPUTING
(2022)
Article
Construction & Building Technology
Sin Yong Tan et al.
Summary: This paper proposes a high-performing and transferable occupancy detection framework that combines sensor data from different data modalities. It also introduces a new metric for evaluating and penalizing delayed occupancy predictions. The algorithms' performance is validated through experiments and analyzed in detail.
ENERGY AND BUILDINGS
(2022)
Article
Engineering, Mechanical
Slah Drira et al.
Summary: This paper presents a non-intrusive and inexpensive method for occupant detection and tracking in buildings using floor-vibration measurements. The method outperforms existing threshold-based methods and successfully distinguishes footsteps from spurious events. It uses support vector-machine classifiers and structural-mechanics models to detect occupants and track their movements. The framework has been tested and validated on two full-scale case studies, demonstrating its utility for buildings with sparse sensor configurations measuring floor vibrations.
MECHANICAL SYSTEMS AND SIGNAL PROCESSING
(2022)
Article
Chemistry, Analytical
Lorenzo Monti et al.
Summary: This paper introduces an approach for people counting using cameras and Raspberry Pi platforms, with specific image processing strategies and an edge-based transfer learning framework. The system was tested in university classrooms and demonstrated feasibility with a maximum mean absolute error of 1.23.
Article
Construction & Building Technology
Abdullah Alsalemi et al.
Summary: Due to the popularity and maturity of AI, Edge Computing has become a popular alternative to cloud platforms in IoE and IoT applications. This paper proposes a novel Edge IoE system called M2SP-EdgeIoE, which incorporates various components such as sensing unit, energy disaggregation unit, anomaly detection platform, and recommender system unit to enable potential IoE use scenarios for domestic energy efficiency applications. The system has achieved promising performance in several scenarios.
SUSTAINABLE CITIES AND SOCIETY
(2022)
Article
Construction & Building Technology
Shuangyu Wei et al.
Summary: This study investigated the potential of real-time occupancy detection approach in demand-controlled ventilation systems, utilizing RCNN models to train people and activity detection, aiming to improve indoor air quality and reduce energy consumption. Experimental results showed that the method could generate count-based occupancy profiles based on real-time information to enhance indoor air quality.
JOURNAL OF BUILDING ENGINEERING
(2022)
Article
Construction & Building Technology
Zeynep Duygu Tekler et al.
Summary: This paper proposes a novel IoT-based plug load management system called Plug-Mate, which reduces energy consumption and user burden through intelligent automation. Different control strategies were evaluated and the optimal balance between automation and user control was found.
BUILDING AND ENVIRONMENT
(2022)
Article
Green & Sustainable Science & Technology
Abolfazl Mohammadabadi et al.
Summary: Indoor occupancy prediction is crucial for the energy-efficient operation of building engineering systems and maintaining satisfactory indoor climate conditions. This study proposes a deep learning method using indoor climate data to predict occupancy and compares its performance with other machine learning algorithms. The results show that the proposed method outperforms other algorithms in predicting occupancy levels in all rooms of the test building.
Article
Computer Science, Interdisciplinary Applications
Ala'a Al-Habashna et al.
Summary: Nowadays, there is a growing interest in indoor localization for mobile applications, which require the location of the devices to operate properly. A study shows that indoor localization and building occupancy-count estimation using 5G Ultra-Dense Networks (UDNs) can significantly improve building operation and management efficiency. This research proposes new models and algorithms to collect and utilize Received Signal Strength Indicator (RSSI) data from User Equipments (UEs) for indoor localization and occupancy-count estimation in buildings.
SIMULATION MODELLING PRACTICE AND THEORY
(2022)
Article
Green & Sustainable Science & Technology
Shahira Assem Abdel-Razek et al.
Summary: Room occupancy prediction based on indoor environmental quality is crucial for energy efficiency and interior design. This research evaluated the accuracy of room occupancy recognition using different datasets. The results showed that KNN performed the best among the classification models tested. By using SHAP, the interpretability of the models was improved.
Review
Automation & Control Systems
Aya Nabil Sayed et al.
Summary: The building internet of things (BIoT) is a promising concept for reducing energy consumption, cutting costs, and promoting building transformation. Integrating artificial intelligence (AI) into the BIoT is crucial for data analysis and intelligent decision-making. This article provides an in-depth survey of strategies used to analyze sensor data and determine building occupancy, with a focus on deep learning and transfer learning approaches. Privacy and precision concerns in the current occupancy detection system are thoroughly discussed. Various directions are proposed to address privacy issues and improve detection accuracy.
ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE
(2022)
Article
Computer Science, Information Systems
Aya Sayed et al.
Summary: The paper proposes integrating an energy efficiency framework into the Home-Assistant platform to achieve energy savings, delivering explainable energy-saving recommendations to users via a mobile application to facilitate changes in energy-saving habits.
IEEE SYSTEMS JOURNAL
(2022)
Article
Green & Sustainable Science & Technology
Anisha Natarajan et al.
Summary: This paper surveys various occupancy detection and localization schemes for smart scheduling of domestic appliances based on user demand in home energy management system. It evaluates the suitability of these schemes on the basis of different factors and suggests feasible solutions. The study finds that wireless technologies and WiFi access points are advantageous for occupancy detection and localization, and proposes a solution that combines them with smartphone inertial sensors and other passive detection schemes.
RENEWABLE & SUSTAINABLE ENERGY REVIEWS
(2022)
Article
Construction & Building Technology
Zeynep Duygu Tekler et al.
Summary: This study performs occupancy prediction based on a minimum sensing strategy using a comprehensive set of sensor data and a proposed feature selection algorithm. The findings highlight the crucial features for predicting occupancy across all space types as indoor CO2 levels and Wi-Fi connected devices.
BUILDING AND ENVIRONMENT
(2022)
Article
Energy & Fuels
Shoaib Azizi et al.
Summary: The advancements in sensor and communication technologies have led to rapid growth in applications for occupancy and indoor environmental monitoring in buildings. However, inadequate sensor positioning can affect data reliability, especially in multi-sensor devices. This study found that the positions of PIR and CO2 sensors significantly impact occupancy detection, highlighting the importance of proper sensor placement for accurate data collection.
Article
Engineering, Electrical & Electronic
Abdullah Alsalemi et al.
Summary: The Internet of Energy is revolutionizing the building energy industry through innovations in data collection, interpretation, and behavioral improvement. Micro-moment analysis helps extract relevant energy consumption data, leading to behavior improvement through recommender systems.
IEEE SENSORS JOURNAL
(2021)
Article
Construction & Building Technology
Chenli Wang et al.
Summary: This paper introduces a cost-effective approach to occupancy detection using a two-layer detection scheme based on data obtained from multiple non-intrusive sensors. Machine learning is utilized for data fusion, enhancing the validity and reliability of occupancy detection. The proposed system shows significant improvements in accuracy and F1-score compared to the current state-of-the-art approach.
ENERGY AND BUILDINGS
(2021)
Article
Multidisciplinary Sciences
Margarite Jacoby et al.
Summary: This paper details the development of a data acquisition system and dataset capturing various occupancy-related modalities in single-family residences. The dataset includes grayscale images, processed audio files, environmental readings, and occupancy status. The system was deployed in six homes for at least one month, capturing data simultaneously from multiple locations within each home.
Article
Engineering, Electrical & Electronic
Florenc Demrozi et al.
Summary: Detecting the presence of persons and estimating their quantity in an indoor environment has become increasingly important. Existing solutions rely on costly hardware installations, while this article proposes a low-cost occupancy detection system, utilizing Bluetooth Low Energy signals to improve accuracy in identifying occupancy.
IEEE SENSORS JOURNAL
(2021)
Article
Tabbi Wilberforce et al.
Energy and Built Environment
(2021)
Proceedings Paper
Engineering, Electrical & Electronic
Tran Ngoc Thinh et al.
Summary: Building smart cities is a highly desired goal in the digital era, with Smart Parking emerging as a core component that promises to automate the parking process, save time and resources, reduce traffic congestion and population density. Utilizing AI/ML/DL algorithms on low-cost System-on-Chip platforms for real-time parking occupancy identification in Smart Parking systems shows promising results with high accuracy, low latency, and high frame per second rate.
2021 INTERNATIONAL CONFERENCE ON ADVANCED TECHNOLOGIES FOR COMMUNICATIONS (ATC 2021)
(2021)
Article
Construction & Building Technology
Zeynep Duygu Tekler et al.
BUILDING AND ENVIRONMENT
(2020)
Review
Construction & Building Technology
Luis Rueda et al.
BUILDING AND ENVIRONMENT
(2020)
Article
Engineering, Civil
Raymond Low et al.
TRANSPORTATION RESEARCH RECORD
(2020)
Article
Engineering, Electrical & Electronic
Cong Feng et al.
IEEE TRANSACTIONS ON SMART GRID
(2020)
Article
Computer Science, Information Systems
Keyan Cao et al.
Article
Construction & Building Technology
Zhe Wang et al.
BUILDING AND ENVIRONMENT
(2019)
Article
Chemistry, Analytical
Bruno Abade et al.
Proceedings Paper
Computer Science, Theory & Methods
Andres Gomez et al.
2018 ACM INTERNATIONAL CONFERENCE ON COMPUTING FRONTIERS
(2018)
Article
Construction & Building Technology
Luis M. Candanedo et al.
ENERGY AND BUILDINGS
(2016)
Proceedings Paper
Computer Science, Theory & Methods
Wilhelm Kleiminger et al.
PROCEEDINGS OF THE 2015 ACM INTERNATIONAL JOINT CONFERENCE ON PERVASIVE AND UBIQUITOUS COMPUTING (UBICOMP 2015)
(2015)
Article
Biochemical Research Methods
Daniel J. Stekhoven et al.