Figure 9 - available via license: Creative Commons Attribution 4.0 International
Content may be subject to copyright.
Source publication
Infrared thermography is the science of measuring the infrared energy emitted by an object, translating it to apparent temperature variance, and displaying the result as an infrared image. Significantly, acquiring thermal images delivers distinctive levels of temperature differences in solar panels that correspond to their health status, which is b...
Contexts in source publication
Context 1
... 5 presents the hyper-parameters which were manually assigned and implemented in our experiments. Figure 9 shows the legend of the confusion matrix for the classification. Below are the following equations for the confusion matrix interpretation. ...
Context 2
... the EL image profile of the solar panels used in the experiment is shown in Figure 18. Figure 19 shows the location of the cell defects on panel 2 for both the thermal and electroluminescence images It shows that the CDF of the sample in the simulated experiments can be validated using the EL images in determining the health status of the solar panel/cell. Although EL is good at detecting defects, it is very costly to set up, which is mostly used in the laboratory. ...
Similar publications
The Rough rice grading is a set of standard methods and processes for evaluating quality that are important to marketing and process controls. IR thermography uses the electromagnetic spectrum’s infrared region to detect heat that has been emitted. FLIR ONE thermal camera was used to acquire images of the sample. In order to represent each sample,...
The increasing number of patients with hypertension is a problem worldwide. Daily blood pressure monitoring is important for the prevention of this disease, and a non‐contact measurement method is desirable. A previous study applied independent components analysis to facial skin temperatures measured remotely using a thermography device for extract...
The aim of the study was to evaluate the temperature parameter of the breast area in patients undergoing radiotherapy at various intervals. The relationship between temperature changes on the patient’s skin and the time after the end of radiotherapy was studied. Measurements with a thermal imaging camera were performed in a group of twelve voluntee...
This paper presents the results of a study on the use of infrared thermography to assess the quality of liquid metal, a basic semi-finished product used in foundry production. EN AC-46000 alloy with the designation AlSi9Cu3(Fe) was used for the study. The crystallization process of the alloy was investigated using the TDA method with a Crystaldigra...
Appropriate matching of rider–horse sizes is becoming an increasingly important issue of riding horses’ care, as the human population becomes heavier. Recently, infrared thermography (IRT) was considered to be effective in differing the effect of 10.6% and 21.3% of the rider:horse bodyweight ratio, but not 10.1% and 15.3%. As IRT images contain man...
Citations
... However, the number of studies on AI-based panel detection using image preprocessing techniques is limited. Although not as comprehensive as the proposed study, the studies in the literature on panel detection using the HE technique are available [29]. present an algorithm that extracts features and simplifies the analysis of thermal images for solar panel applications. ...
... Despite the large dataset, the F1 score value was 14 % lower than the proposed study. Although a dataset about three times higher than the current work used in the [29], however similar performance rates could only be achieved by applying the HE preprocessing technique. Moreover, while three different classes were detected in the proposed research, two classes were detected in [29]. ...
... Although a dataset about three times higher than the current work used in the [29], however similar performance rates could only be achieved by applying the HE preprocessing technique. Moreover, while three different classes were detected in the proposed research, two classes were detected in [29]. The 97 % success rate obtained in the proposed study at [30] could only be achieved with a dataset two times larger than the dataset used in the proposed study. ...
... Grayscale has a range of color gradations between black and white which is very appropriate for image processing [16]. Image conversion to grayscale is performed to clarify contrast variations in the image [27]. This grayscale process can be described by the Equation (1) [28]. ...
Diabetic Retinopathy (DR) is a disorder of the eye caused by damage to blood vessels in the retina. Damage to the retinal blood vessels can be analyzed by segmenting the blood vessels on the image. This study proposes a combination of image enhancement and blood vessel segmentation in retinal images. Retinal image enhancement is carried out using the black hat transform method to obtain a detailed view of blood vessels in retinal images. Segmentation of blood vessels in retinal images is carried out using the U-Net architecture. The results of image enhancement are measured using MSE and PSNR. This study has an MSE value below 0.05 and a PSNR above 90dB. The MSE and PSNR values obtained show that the black hat transform method is very good at image enhancement. Segmentation has an accuracy value above 0.95 and a sensitivity value above 0.85. In addition, the specificity value and f1-score are above 0.8. This shows that the proposed stages of image enhancement and blood vessel segmentation are able to accurately recognize blood vessel features in retinal images.
... Several cases of fire caused by PV systems were reported and investigated [17][18][19]. A local temperature rise caused these fires, called hot spots, and their temperature rise was the concentration of the energy loss of the solar cells owing to current-mismatching loss [20][21][22][23][24]. Thermal images typically monitor and detect these local losses [25][26][27][28][29][30][31][32][33]. However, it is essential to detect early symptoms using non-contact and non-destructive test methods and to quantify the modeling and measurement of the degree of concentration of the mismatching loss, for example, xx W loss (xx may be obtained by measurement) in this particular solar cell. ...
Fires in photovoltaic modules are caused by hot spots, which are typically monitored by thermal images. This method helps visualize the hot spot, but it is affected by the environment (solar irradiance, wind, ambient temperature) and is not reproducible. Assessing the heat dissipation of the hot cell can be used for alternative assessment of the fire risk. This method was validated by comparing the value measured by the surface potential meter and the module potential measured directly by adding a bypass measurement circuit. The substantial reverse-bias voltage caused by mismatching or partial shading (depending on the operating conditions) leads to local heat consumption of the partially shaded solar cells and potentially causes fire. The fire risk can be assessed in the worst-case conditions (ex. 1380 W/m2 solar irradiance) by non-contact measurement of the reverse-bias voltage and calculating the heat dissipation and temperature rise. This work suggested that −13 V is the criterion and was close to the known value of reverse voltage for Si cells. The current technology inspects solar cells before assembly to the module, and there is no way of inspecting in the product test or detecting after degradation that can be covered by the proposed method in this work.
... In [13], the authors combined a CNN with a support vector machine to classify the IRT images of PV modules. In addition, the authors used both the image histogram information and its corresponding CDF (cumulative distribution function). ...
In this paper, a novel hybrid model for the fault diagnosis of photovoltaic (PV) modules was developed. The model combines a convolutional neural network (CNN) with a machine learning (ML) algorithm. A total of seven defects were considered in this study: sand accumulated on PV modules, covered PV modules, cracked PV modules, degradation, dirty PV modules, short-circuited PV modules, and overheated bypass diodes. First, the hybrid CNN–ML has been developed to classify the seven common defects that occur in PV modules. Second, the developed model has been then optimized. Third, the optimized model has been implemented into a microprocessor (Raspberry Pi 4) for real-time application. Finally, a friendly graphical user interface (GUI) has been designed to help users analyze their PV modules. The proposed hybrid model was extensively evaluated by a comprehensive database collected from three regions with different climatic conditions (Mediterranean, arid, and semi-arid climates). Experimental tests showed the feasibility of such an embedded solution in the diagnosis of PV modules. A comparative study with the state-of-the-art models and our model has been also presented in this paper.
... A beneficial technique for identifying the abnormalities in photovoltaics is infrared, which operates a thermal IR camera to display faults depending on their raised heat [3]. A thermal camera may benefit from showing unsafe parts emitting temperatures over their standard operating boundary and the high temperature of photovoltaic cells related to anomalies in the photovoltaic system [4]. For example, surface imperfections result in heated cells called hot spots on the photovoltaic module. ...
... The line illustrates the probability using vertical distances. While a histogram is more straightforward when assessing image processing, the CDF is a more practical idea [4]. Slopes represent data in CDF, as seen in Table 2. Our eyes are significantly more robust at detecting slope variations, making it sometimes simpler to spot outliers in the sample. ...
The condition of photovoltaic thermal image data is crucial to a great variety of developing research and implementations since thermal images are competent in exposing meaningful unseen features for detecting hotspots as the abnormality in a pho-tovoltaic module. However, the images may degrade from unsatisfactory image quality because several images may have too low contrast. In addition, the details are at variance with human visual perception due to limitations of the image acquisition device and di-versities of the surroundings of the photovoltaic. Therefore, an image enhancement stage is needed to solve these issues. This research presents the comparison methods for the image enhancement stage that promotes the details of photovoltaic thermal images by enhancing contrast. These methods were completed using contrast-stretching and histogram equalization. The findings indicated that histogram equalization gave superior histogram and CDF properties than contrast-stretching. Additionally, it recognized hotspot cells well in quantitative analysis, with an average accuracy of 96.93% and an average F1 Score of 81.84% from 30 photovoltaic thermal images.
... Furthermore, only few authors addressed the topic of "in falso" walls [28,29,30] and, for this reason, an innovative experimental setup is proposed, aimed at investigating the features of such masonry elements that in case of seismic action behave like "wall beams". In addition, considering that innovative interventions are presented for high value architectures, different approaches to data acquisition techniques and monitoring were also briefly analysed that are both respectful of the environment and of the heritage [31,32,33,34,35,36,37,38]. Finally, it is foreseen, as a future development of the ongoing research, that the experimental campaign and its evolution could be the basis for the creation of a proper numerical modelling strategy. ...
The present paper contains some ‘spoilers’ concerning some of the issues addressed in the PhD thesis of one of the authors, Riccardo Liberotti, except for the modelling: indeed, he developed in its period at the Universidade do Minho, under the guidance of Prof. P. Lourenço, a different numerical solution which will be the subject of future publications.
Here, however, the work is focused on an experimental campaign and a preliminary and alternative modelling – that is now being refined with the other authors – concerning the preservation of heritage architectures: not just about the structural safety but necessarily related to the central themes of restoration, fruition and reuse of ancient buildings.
These topics require an interdisciplinary design approach that involves, among the others, structural engineering, numerical modelling and architecture to address the challenges of contemporaneity in heritage management also in terms of interests of the stakeholders. In this regard, the opportunities offered by natural F.R.C.M. (Fibre Reinforced Cementitious Matrix) composites, made of basaltic fibres and lime mortar, are analysed. The main objective of the research is to expand the state of the art concerning the influence of such composites’ applications on «in falso» masonries: load bearing walls built without a direct load path to the ground and acting like «wall beams» in case of collapse of the underlying masonry vaults. On this, an experimental campaign started in order to make a comparative assessment about the influence of composites in the prevention of knock-on collapses due to the aforementioned scenario, with the benefit of avoiding damages to unmovable artistic assets and valuable architectural features (e.g. precious pavings, mosaics, vaulted surfaces etc.). Furthermore, rather than the «canon» widespread intervention, an innovative application of F.R.C.M. in tapes is proposed, aimed at interventions with reduced quantities of composite material and therefore improving its use in the heritage even more. Finally, a first approach to the 3D Finite Element formulation for the case in question is proposed.
... This image consists of noise and is unclear due to different environmental and technical parameters. From the previous studies, it has been concluded that by using adaptive histogram equalization grayscale and histogram equalization, the thermal solar panel images can be enhanced [64]. Moreover, with the different models such as artificial neural network (ANN) and convolutional neural network (CNN), the support vector machine achieved the highest accuracy in identifying the hotspot of faults in the enhanced images [64][65][66]. ...
... From the previous studies, it has been concluded that by using adaptive histogram equalization grayscale and histogram equalization, the thermal solar panel images can be enhanced [64]. Moreover, with the different models such as artificial neural network (ANN) and convolutional neural network (CNN), the support vector machine achieved the highest accuracy in identifying the hotspot of faults in the enhanced images [64][65][66]. The above facts will encourage the implementation of the same models in the field of BIPV. ...
Building integrated photovoltaic (BIPV) systems have gained a lot of attention in recent years as they support the United Nations’ sustainable development goals of renewable energy generation and construction of resilient infrastructure. To make the BIPV system infra resilient, there is a need to adopt digital technologies such as the internet of things (IoT), artificial intelligence (AI), edge computing, unmanned aerial vehicles (UAV), and robotics. In this study, the current challenges in the BIPV system, such as the rise in the temperature of the PV modules, the occurrence of various faults, and the accumulation of dust particles over the module surface, have been identified and discussed based on the previous literature. To overcome the challenges, the significance and application of the integration of these digital technologies in the BIPV system are discussed along with the proposed architecture. Finally, the study discusses the vital recommendations for future directions, such as ML and DL for image enhancement and flaws detection in real-time image data; edge computing to implement DL for intelligent BIPV data analytics; fog computing for 6G assisted IoT network in BIPV; edge computing integration in UAV for intelligent automation and detection; augmented reality, virtual reality, and digital twins for virtual BIPV systems with research challenges of real-time implementation in the BIPV.
... Furthermore, only a few authors have addressed the topic of "in falso" walls [29,30] and, for this reason, an innovative experimental setup is proposed in this paper, aimed at investigating their structural behavior and possible intervention strategies for their reinforcement. In addition, considering that the interventions proposed in the paper are designed for architectural heritage, different approaches and data acquisition techniques are also briefly analyzed that are both respectful of the environment and the architectural value of historical architecture [31][32][33][34][35][36]. Finally, it is then foreseen, as a future development of such ongoing research, that the experimental campaign and its evolution could be the basis for the creation of a proper numerical modeling strategy. ...
In this paper, an innovative strengthening system for masonry walls made of externally bonded Fabric-Reinforced Cementitious Matrix (FRCM) is presented. Due to the good mechanical properties and the compatibility with the architectural heritage, the FRCM is an adequate alternative to the use of Fiber-Reinforced Polymer (FRP) composites and other traditional techniques. The proposed system is applied to the strengthening of a classical architectural typology in cultural heritage architecture, which is the “in falso” masonry: a load-bearing wall built over a masonry vault, and hence without a direct load path to the ground. A research program, characterized by an experimental campaign, has been started in order to devise and verify an optimal strengthening system that assures for the masonry wall a structural behavior similar to a “wall beam”, so to prevent progressive collapses when the underlying masonry vault loses its carrier function. In particular, rather than the canonical application, consisting in widespread application to the whole surface of the masonry wall, an innovative intervention made of “Green Tape” of composites has been designed and verified by a specifically designed experimental set-up. The main objective of the research is to propose a reinforcement strategy not detrimental to unmovable artistic assets and tied to the safety and robustness of the architectural heritage.
... c) Otsu's Method -creates a threshold for converting the grayscale image into a binary image in which every pixel only represents 0 and 1. This algorithm is required before proceeding into any type of edge detection algorithms [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26]. d) Canny Edge Detection -the algorithm for taking the outline of the flood line which is necessary for the calculations [27], [28]. ...
In the Philippines, flooding, which is typically produced by excessive rainfall and strong seas, is one of the most prevalent natural occurrences. This natural calamity cannot be avoided but the good thing is, we can practice ourselves to be prepared for it. After conducting an analysis regarding the needs of people residing in Barangay Frances, Calumpit, Bulacan, it was then decided to develop a project that can help lessen the difficulty they are experiencing when they evacuate. The system uses image processing as its flood detection method. It also uses several sensors for different purposes to make it more reliable to the users. These sensors used are the rain gauge, float switch, and flow rate meter sensor. It measures two of the important parameters in flood detection namely precipitation rate (mm/hr), flood level (ft), and the flow rate (L/hr). The data accumulated by the sensors are sent immediately to the Android application so it can be used by people living in the area to monitor the flood levels in real time. To measure the reliability of the system, the flood level taken from the automated system and conventional method were compared. A small mean squared error (MSE) of these 2 data which is 0.125 was achieved.
... Therefore, recent applications of CNN-based models for SD assessments focus more on quality data preparation, the algorithmic optimization of the CNN model architecture, and damage quantification. Techniques typically adopted for quality image data preparation include image enhancement approaches, such as gray-level thresholding, histogram equalization, and adaptive histogram equalization [20]. Moreover, the algorithmic optimization of hyperparameters enhances the accuracy of CNN-based models and reduces the computational power used for execution [21]. ...
The adoption of artificial intelligence in post-earthquake inspections and reconnaissance has received considerable attention in recent years, owing to its exponential increase in computation capabilities and inherent potential in addressing disadvantages associated with manual inspections. Herein, we present the effectiveness of automated deep learning in enhancing the assessment of damage caused by the 2017 Pohang earthquake. Six classical pre-trained convolutional neural network (CNN) models are implemented through transfer learning (TL) on a small dataset, comprising 1780 manually labeled images of structural damage. Feature extraction and fine-tuning TL methods are trained on the image datasets. The performances of various CNN models are compared on a testing image dataset. Results confirm that the MobileNet fine-tuned model offers the best performance. Therefore, the model is further developed as a web-based application for classifying earthquake damage. The severity of damage is quantified by assigning damage assessment values, derived using the CNN model and gradient-weighted class activation mapping. The web-based application can effectively and automatically classify structural damage resulting from earthquakes, rendering it suitable for decision making, such as in resource allocation, policy development, and emergency response.
