Real-time surveillance-video-based personalized thermal comfort recognition

Duan, Wenjun; Wang, Yu; Li, Junqing; Zheng, Yuanjie; Ning, Chenguang; Duan, Peiyong

doi:10.1016/j.enbuild.2021.110989

Cited by 26 publications

(6 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In addition to using thermal imaging cameras to obtain personal information, many studies use cameras to obtain image data and combine different image recognition to obtain personal information. Peixian Li et al [9] used the CNN model to identify the gender and age of people, Songtao Hu et al [10] evaluated the thermal comfort of the current environment by recognizing facial expressions, Jinsong Liu et al [11] And Haneul Choi et al [12] identify the clothing (5 types) and actions (standing and walking) of the personnel and then calculate the PMV value, Wenjun Duan et al [13] conducted further research on the relationship between action recognition (16 types) and thermal comfort. Eun Ji Choi et al [14] identified 16 different kinds of clothing and compared the thermal comfort and power consumption of three different air-conditioning control modes.…”

Section: Pmv Measurement Methodsmentioning

confidence: 99%

Development of personalized predicted mean vote based on a real-time clothing insulation recognition system

Shih,

Chan

2023

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

As people’s standards for indoor environmental quality and life are getting higher and higher, many studies try to find ways to improve indoor thermal comfort through experiments and simulations. Predictive mean voting (PMV) is a widely used indicator, but personal factors used to calculate PMV are not easy to monitor. This study aimed to develop a real-time clothing insulation identification (R-CLO) system to estimate personal PMV. This study is mainly divided into three stages. The first stage is to issue questionnaires to obtain the value of the personal clothing level and analyze the correlation between different external temperatures and the clothing insulation (Icl). The second stage is to build a real-time clothing recognition model and analyze its recognition accuracy. The third stage is to use ANN to establish a clothing insulation transition model and use the outdoor temperature and the results of clothing recognition to estimate Icl. This research also established an experimental environment to verify the system’s feasibility. The results show that the system has more than 80% accuracy rate for clothes recognition, and the accuracy rate of short-sleeved tops, shorts, trousers, and skirts is more than 90%. Moreover, most of the Icl estimated by the ANN have less than 0.1 differences from the value reported by occupants. In the future, this system can also combine occupant behavior recognition to strengthen PMV prediction results to achieve more precise, comfortable air-conditioning control.

show abstract

Section: Pmv Measurement Methodsmentioning

confidence: 99%

Development of personalized predicted mean vote based on a real-time clothing insulation recognition system

Shih,

Chan

2023

J. Phys.: Conf. Ser.

View full text Add to dashboard Cite

show abstract

“…Sensing research is in its early stages at present and focuses on using thermal adaptation behavior to recognize actions by analyzing skeleton-based information. The research in this field may be divided into two categories: manually constructed algorithms that rely on features [ 38 , 39 ] and algorithms that use spatial–temporal graph convolutional networks [ 40 ]. Algorithms that depend on manually designed features obtain visual input using computer cameras or Kinect sensor devices.…”

Section: Related Workmentioning

confidence: 99%

“…A dataset including films from thermal comfort research initiatives was uploaded; we specifically focused on thermal adaptation actions (TAAs) [ 40 ]. The dataset had around 14,800 movies that are deemed legitimate and that showcase 16 distinct temperature adaptation behaviors.…”

Section: Model Evaluationmentioning

confidence: 99%

“…Furthermore, individuals may not be consistently positioned in front of a computer screen and may sometimes be beyond the camera’s range. To address the limitations of the previously described approaches, a study [ 40 ] proposed a thermal adaption behavior detection technique that utilizes spatial–temporal graph convolutional networks (ST-GCNs) [ 41 ]. The researchers used OpenPose and spatial–temporal graph convolutional networks to construct a model to identify thermal adaptive behavior.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Thermal-Adaptation-Behavior-Based Thermal Sensation Evaluation Model with Surveillance Cameras

Wang,

Duan,

et al. 2024

Sensors

Self Cite

View full text Add to dashboard Cite

The construction sector is responsible for almost 30% of the world’s total energy consumption, with a significant portion of this energy being used by heating, ventilation and air-conditioning (HVAC) systems to ensure people’s thermal comfort. In practical applications, the conventional approach to HVAC management in buildings typically involves the manual control of temperature setpoints by facility operators. Nevertheless, the implementation of real-time alterations that are based on the thermal comfort levels of humans inside a building has the potential to dramatically improve the energy efficiency of the structure. Therefore, we propose a model for non-intrusive, dynamic inference of occupant thermal comfort based on building indoor surveillance camera data. It is based on a two-stream transformer-augmented adaptive graph convolutional network to identify people’s heat-related adaptive behaviors. The transformer specifically strengthens the original adaptive graph convolution network module, resulting in further improvement to the accuracy of the detection of thermal adaptation behavior. The experiment is conducted on a dataset including 16 distinct temperature adaption behaviors. The findings indicate that the suggested strategy significantly improves the behavior recognition accuracy of the proposed model to 96.56%. The proposed model provides the possibility to realize energy savings and emission reductions in intelligent buildings and dynamic decision making in energy management systems.

show abstract

“…Our method is based on thermal imaging, which has been widely used in medicine ( Gizińska et al, 2021 ; Martinez-Jimenez et al, 2021 ), surveillance ( Ivašić-Kos et al, 2019 ; Duan et al, 2021 ), and quality control ( ElMasry et al, 2020 ; Khera et al, 2020 ). The starting point of our application is the temperature of fingers, which is 29.1 ± 0.6°C at the environmental temperature of 25.4 ± 0.4°C, ( Shilco et al, 2019 ), and which transfers to objects that have been touched.…”

Section: Introductionmentioning

confidence: 99%

A Method for Measuring Contact Points in Human–Object Interaction Utilizing Infrared Cameras

Hakala

Häkkinen

2022

Front. Robot. AI

View full text Add to dashboard Cite

This article presents a novel method for measuring contact points in human–object interaction. Research in multiple prehension-related fields, e.g., action planning, affordance, motor function, ergonomics, and robotic grasping, benefits from accurate and precise measurements of contact points between a subject’s hands and objects. During interaction, the subject’s hands occlude the contact points, which poses a major challenge for direct optical measurement methods. Our method solves the occlusion problem by exploiting thermal energy transfer from the subject’s hand to the object surface during interaction. After the interaction, we measure the heat emitted by the object surface with four high-resolution infrared cameras surrounding the object. A computer-vision algorithm detects the areas in the infrared images where the subject’s fingers have touched the object. A structured light 3D scanner produces a point cloud of the scene, which enables the localization of the object in relation to the infrared cameras. We then use the localization result to project the detected contact points from the infrared camera images to the surface of the 3D model of the object. Data collection with this method is fast, unobtrusive, contactless, markerless, and automated. The method enables accurate measurement of contact points in non-trivially complex objects. Furthermore, the method is extendable to measuring surface contact areas, or patches, instead of contact points. In this article, we present the method and sample grasp measurement results with publicly available objects.

show abstract

Real-time surveillance-video-based personalized thermal comfort recognition

Cited by 26 publications

References 32 publications

Development of personalized predicted mean vote based on a real-time clothing insulation recognition system

Development of personalized predicted mean vote based on a real-time clothing insulation recognition system

Thermal-Adaptation-Behavior-Based Thermal Sensation Evaluation Model with Surveillance Cameras

A Method for Measuring Contact Points in Human–Object Interaction Utilizing Infrared Cameras

Contact Info

Product

Resources

About