scholarly journals Automatic Detection of Linear Thermal Bridges from Infrared Thermal Images Using Neural Network

2021 ◽  
Vol 11 (3) ◽  
pp. 931
Author(s):  
Changmin Kim ◽  
Jae-Sol Choi ◽  
Hyangin Jang ◽  
Eui-Jong Kim
Keyword(s):  
Neural Network ◽  
Detection Method ◽  
Early Stage ◽  
Thermal Anomaly ◽  
Building Envelopes ◽  
Subjective Judgment ◽  
Thermal Images ◽  
Thermal Bridge ◽  
Thermal Bridges ◽  
Infrared Thermal Images

Detecting thermal bridges in building envelopes should be a priority to improve the thermal performance of buildings. Recently, thermographic surveys are being used to detect thermal bridges. However, conventional methods of detecting thermal bridges from thermal images rely on the subjective judgment of audits. Research has been conducted to automatically detect thermal bridges from thermal images to improve problems caused by such subjective judgment, but most of these studies are still in the early stage. Therefore, this study proposes a linear thermal bridge detection method based on image processing and machine learning. The proposed method includes thermal anomaly area clustering, feature extraction, and an artificial-neural-network-based thermal bridge detection. The proposed method was validated by detecting the thermal bridges in actual buildings. As a result, the average precision, recall, and F-score were 89.29%, 87.29, and 87.63%, respectively.

Comparison of Thermal Bridges Calculate Method through Window Jamb

2013 ◽  
Vol 855 ◽  
pp. 130-133
Author(s):  
Rastislav Ingeli
Keyword(s):  
Heat Conduction ◽  
Thermal Conductance ◽  
Theoretical Background ◽  
Building Envelope ◽  
Building Construction ◽  
Building Envelopes ◽  
Thermal Bridge ◽  
Detailed Specification ◽  
Flow Through ◽  
Thermal Bridges

This paper is focused on comparison of thermal bridges calculate method through window jamb in building envelopes. The present approach is based on an integrated 2D dynamic simulation. The theoretical background of the adopted approach is presented. The reliability of this approach in evaluating thermal bridges as well as its applicability to different geometric shapes is proved. Detailed specification and calculation of each thermal bridge in these buildings should be taken into account. the heat flow through a building construction is considered to be of the onedimensional (1D) type. This is because the thermal conductance and temperature differential in this direction are much greater than that in the lateral directions. The thermal bridge is the part of the building envelope through which heat conduction is multi-dimensional. Therefore, in recent studies, the problem of heat conduction in the building construction has been treated as a multi-dimensional.

Thermal Bridges Minimizing through Window Jamb in Low Energy Buildings

2014 ◽  
Vol 899 ◽  
pp. 66-69 ◽  
Author(s):  
Rastislav Ingeli ◽  
Boris Vavrovič ◽  
Miroslav Čekon ◽  
Lucia Paulovičová
Keyword(s):  
Thermal Protection ◽  
Low Energy ◽  
Building Envelopes ◽  
Thermal Bridge ◽  
Window Systems ◽  
Significant Difference ◽  
Low Energy Buildings ◽  
Window Position ◽  
High Thermal Resistance ◽  
Thermal Bridges

Building envelopes with high thermal resistance are typical for low-energy buildings. Detailed specification and calculation of each thermal bridge in these buildings should be taken into account. This paper is focused on thermal bridges minimizing through typical window systems in building envelopes. The aim of this article is to analyze the window position influence, as regards on thermal performance and to point out the installation modality in accordance with the characterization of the windows performance. This can be done by quantifying the percentage increment of the window jamb thermal transmittance. The calculated results also demonstrate that there is significant difference between results obtained by various available calculation approaches. This can be significant especially in buildings with high thermal protection.

scholarly journals Instantaneous field measurements of thermal bridge parameters in ground floor residential room

2019 ◽  
Vol 112 ◽  
pp. 01016 ◽  
Author(s):  
Martin Ivanov
Keyword(s):  
Energy Demand ◽  
Field Measurements ◽  
Building Design ◽  
Building Envelope ◽  
Ground Floor ◽  
Thermal Bridge ◽  
Significant Temperature ◽  
Thermal Bridges ◽  
Infrared Thermal Images ◽  
Heating Energy Demand

The “thermal bridges” are defined as an isolated building’s areas, where the construction elements have higher thermal conductivity, compared with the rest of the building envelope. Thus, at cold winter conditions, a significant temperature difference may occur between neighbouring solid and air volumes within the construction. Moreover, it has been documented, that the heating energy demand of a building may be increased with more than 30%, due to the existence of thermal bridges and the increased heat losses from the indoors. Consequently, in the recent years, norms and standards have been developed, for avoiding thermal bridges during the building design process. But still, thermal bridges exist in the indoor environment, especially in older buildings, where no energy efficient measures have been applied. That is why, the presented study focuses on instantaneous field measurements of thermal bridge parameters in real existing ground floor residential room. The thermal bridge propagation is analysed relative to the indoor and outdoor air temperature and relative humidity, as well as with infrared thermal images of the affected external walls. The achieved results give valuable information about the generic conditions for thermal bridge existence, without considering the building envelope properties.

scholarly journals Heat and moisture transfer in wall-to-floor thermal bridges and its influences on the insulation performance

2021 ◽  
Vol 2069 (1) ◽  
pp. 012216
Author(s):  
Yucong Xue ◽  
Jian Ge ◽  
Yifan Fan
Keyword(s):  
Water Vapour ◽  
Moisture Transfer ◽  
Structural Safety ◽  
Building Envelopes ◽  
Thermal Bridge ◽  
Thermal Bridges ◽  
Heat Transfers ◽  
Heat And Moisture ◽  
Insulation Performance ◽  
Transient Numerical Simulation

Abstract The moisture modifies the characteristics of heat transfer in building envelopes. Multiple factors, including the distinct hygric properties of various material, gravity, etc., affect the moisture content, resulting in a non-uniform distribution of water vapour in different parts of the envelope (e.g. column, beam, the main part of exterior walls). Usually, the more water vapour in a material, the higher the thermal conductivity, resulting in more heat transfers here. Moreover, condensation easily occurs where there is wet, marking such parts have risks both on structural safety and mould growth. The wall-to-floor thermal bridge (WFTB) occupies the largest area among all kinds of thermal bridges that formed by frame structures. In this study, we aimed to quantify the influence on heat loss through WFTB when the moisture transfer in envelopes is considered. The average apparent thermal resistance of WFTB (R TB, ave) was defined to access the insulation performance of WFTB in practical application. The results of transient numerical simulation indicated that when the moisture transfer is considered, the insulation performance of building envelopes decreases significantly, while the adverse effect of WFTB on heat insulation becomes less pronounced. The results indicated that the measures of insulation for WFTB should be reconsidered when the moisture transfer is considered.

On-Site Identification of Zero Resistance Insulator Based on Infrared Thermal Image and Weights-Direct-Determination Neural Network

2013 ◽  
Vol 441 ◽  
pp. 417-420
Author(s):  
Tang Bing Li ◽  
Lei Gong ◽  
Jian Gang Yao ◽  
Yan Jun Kuang ◽  
Bin Bin Rao
Keyword(s):  
Neural Network ◽  
Transmission Lines ◽  
Direct Determination ◽  
Thermal Image ◽  
Absolute Deviation ◽  
Thermal Images ◽  
Zero Resistance ◽  
Infrared Thermal Images ◽  
Site Identification ◽  
The Relationship

A method using infrared thermal images and weights-direct-determination neural network (WDDNN) to identify the zero resistance insulators on-site is presented. The basic procedures were as follows: the infrared thermal image were denoised, intensified, segmented, and a rectangular which was regarded as object was intercepted in the insulators chain; in view of the relationship between gray value of infrared thermal images and temperature of object surface, four parameters which stand for standard deviation, absolute deviation, quartiles and range of gray value, were extracted directly; these four parameters were used as the input of WDDNN to train the model, which could be used identifing the zero resistance insulators after being trained. This method can effectively avoid the interference of transmission lines, and can meet the real-time require when identifying on-site. Experimental results verify the feasibility and effectiveness of this method.

Identification of Coronary Artery Disease using Artificial Neural Network and Case Based Reasoning

2020 ◽  
Vol 13 ◽  
Author(s):  
Varun Sapra ◽  
M.L Saini ◽  
Luxmi Verma
Keyword(s):  
Neural Network ◽  
Coronary Artery Disease ◽  
Coronary Artery ◽  
Early Stage ◽  
Second Phase ◽  
Case Based Reasoning ◽  
Data Set ◽  
Artery Disease ◽  
Severity Of The Disease ◽  
Case Based

Background: Cardiovascular diseases are increasing at an alarming rate with very high rate of mortality. Coronary artery disease is one of the type of cardiovascular disease, which is not easily diagnosed in its early stage. Prevention of Coronary Artery Disease is possible only if it is diagnosed, at early stage and proper medication is done. Objective: An effective diagnosis model is important not only for the early diagnosis but also to check the severity of the disease. Method: In this paper, a hybrid approach is followed, with the integration of deep learning (multi-layer perceptron) with Case based reasoning to design analytical framework. This paper suggests two phases of the study, one in which the patient is diagnosed for Coronary artery disease and in second phase, if the patient is suffering from the disease then employing Case based reasoning to diagnose the severity of the disease. In the first phase, multilayer perceptron is implemented on reduced dataset and with time-based learning for stochastic gradient descent respectively. Results: The classification accuracy is increase by 4.18 % with reduced data set using deep neural network with time based learning. In second phase, if the patient is diagnosed as positive for Coronary artery disease, then it triggers the Case based reasoning system to retrieve from the case base, the most similar case to predict the severity for that patient. The CBR model achieved 97.3% accuracy. Conclusion: The model can be very useful for medical practitioners as a supporting decision system and thus can save the patients from unnecessary medical expenses on costly tests and can improve the quality and effectiveness of medical treatment.

The Convolution Neural Network Combined with the HT Person Fit Statistic to Develop an APP for Detecting Dengue Fever in Children: Development and Usability Study (Preprint)

2019 ◽  
Author(s):  
CHIEN WEI ◽  
Chi Chow Julie ◽  
Chou Willy
Keyword(s):  
Neural Network ◽  
Dengue Fever ◽  
Prediction Accuracy ◽  
Early Stage ◽  
Family Members ◽  
Convolution Neural Network ◽  
Public Health Issue ◽  
Person Fit ◽  
Model Accuracy ◽  
Comparison Results

UNSTRUCTURED Backgrounds: Dengue fever (DF) is an important public health issue in Asia. However, the disease is extremely hard to detect using traditional dichotomous (i.e., absent vs. present) evaluations of symptoms. Convolution neural network (CNN), a well-established deep learning method, can improve prediction accuracy on account of its usage of a large number of parameters for modeling. Whether the HT person fit statistic can be combined with CNN to increase the prediction accuracy of the model and develop an application (APP) to detect DF in children remains unknown. Objectives: The aim of this study is to build a model for the automatic detection and classification of DF with symptoms to help patients, family members, and clinicians identify the disease at an early stage. Methods: We extracted 19 feature variables of DF-related symptoms from 177 pediatric patients (69 diagnosed with DF) using CNN to predict DF risk. The accuracy of two sets of characteristics (19 symptoms and four other variables, including person mean, standard deviation, and two HT-related statistics matched to DF+ and DF−) for predicting DF, were then compared. Data were separated into training and testing sets, and the former was used to predict the latter. We calculated the sensitivity (Sens), specificity (Spec), and area under the receiver operating characteristic curve (AUC) across studies for comparison. Results: We observed that (1) the 23-item model yields a higher accuracy rate (0.95) and AUC (0.94) than the 19-item model (accuracy = 0.92, AUC = 0.90) based on the 177-case training set; (2) the Sens values are almost higher than the corresponding Spec values (90% in 10 scenarios) for predicting DF; (3) the Sens and Spec values of the 23-item model are consistently higher than those of the 19-item model. An APP was subsequently designed to detect DF in children. Conclusion: The 23-item model yielded higher accuracy rates (0.95) and AUC (0.94) than the 19-item model (accuracy = 0.92, AUC = 0.90). An APP could be developed to help patients, family members, and clinicians discriminate DF from other febrile illnesses at an early stage.

Jaya Ant Lion Optimization-Driven Deep Recurrent Neural Network for Cancer Classification Using Gene Expression Data (Preprint)

2020 ◽  
Author(s):  
Ramachandro Majji
Keyword(s):  
Neural Network ◽  
Recurrent Neural Network ◽  
Early Stage ◽  
Data Transformation ◽  
Cancer Classification ◽  
Data Normalization ◽  
Ant Lion Optimization ◽  
Deep Recurrent Neural Network ◽  
Novel Strategy ◽  
Feature Dimension

BACKGROUND Cancer is one of the deadly diseases prevailing worldwide and the patients with cancer are rescued only when the cancer is detected at the very early stage. Early detection of cancer is essential as, in the final stage, the chance of survival is limited. The symptoms of cancers are rigorous and therefore, all the symptoms should be studied properly before the diagnosis. OBJECTIVE Propose an automatic prediction system for classifying cancer to malignant or benign. METHODS This paper introduces the novel strategy based on the JayaAnt lion optimization-based Deep recurrent neural network (JayaALO-based DeepRNN) for cancer classification. The steps followed in the developed model are data normalization, data transformation, feature dimension detection, and classification. The first step is the data normalization. The goal of data normalization is to eliminate data redundancy and to mitigate the storage of objects in a relational database that maintains the same information in several places. After that, the data transformation is carried out based on log transformation that generates the patterns using more interpretable and helps fulfill the supposition, and to reduce skew. Also, the non-negative matrix factorization is employed for reducing the feature dimension. Finally, the proposed JayaALO-based DeepRNN method effectively classifies cancer-based on the reduced dimension features to produce a satisfactory result. RESULTS The proposed JayaALO-based DeepRNN showed improved results with maximal accuracy of 95.97%, the maximal sensitivity of 95.95%, and the maximal specificity of 96.96%. CONCLUSIONS The resulted output of the proposed JayaALO-based DeepRNN is used for cancer classification.

Sign in / Sign up

Export Citation Format

Share Document