Search results for: deep mixing method

Authors: Nouha Khediri, Mohammed Ben Ammar, Monji Kherallah

Abstract:

Recently, facial emotion recognition (FER) has become increasingly essential to understand the state of the human mind. Accurately classifying emotion from the face is a challenging task. In this paper, we present a facial emotion recognition approach named CV-FER, benefiting from deep learning, especially CNN and VGG16. First, the data is pre-processed with data cleaning and data rotation. Then, we augment the data and proceed to our FER model, which contains five convolutions layers and five pooling layers. Finally, a softmax classifier is used in the output layer to recognize emotions. Based on the above contents, this paper reviews the works of facial emotion recognition based on deep learning. Experiments show that our model outperforms the other methods using the same FER2013 database and yields a recognition rate of 92%. We also put forward some suggestions for future work.

Keywords: CNN, deep-learning, facial emotion recognition, machine learning

Procedia PDF Downloads 61

Authors: Seyed Mostafa Jafari Raad, Hassan Hassanzadeh

Abstract:

Injection of impurities along with CO₂ into saline aquifers provides an exceptional prospect for low-cost carbon capture and storage technologies and can potentially accelerate large-scale implementation of geological storage of CO₂. We have conducted linear stability analyses and numerical simulations to investigate the effects of permitted impurities in CO₂ streams on the onset of natural convection and dynamics of subsequent convective mixing. We have shown that the rate of dissolution of an impure CO₂ stream with H₂S highly depends on the operating conditions such as temperature, pressure, and composition of impurity. Contrary to findings of previous studies, our results show that an impurity such as H₂S can potentially reduce the onset time of natural convection and can accelerate the subsequent convective mixing. However, at the later times, the rate of convective dissolution is adversely affected by the impurities. Therefore, the injection of an impure CO₂ stream can be engineered to improve the rate of dissolution of CO₂, which leads to higher storage security and efficiency. Accordingly, we have identified the most favorable CO₂ stream compositions based on the geophysical properties of target aquifers. Information related to the onset of natural convection such as the scaling relations and the most favorable operating conditions for CO₂ storage developed in this study are important in proper design, site screening, characterization and safety of geological storage. This information can be used to either identify future geological candidates for acid gas disposal or reviewing the current operating conditions of licensed injection sites.

Keywords: CO₂ storage, solubility trapping, convective dissolution, storage efficiency

Procedia PDF Downloads 170

Authors: Diana G. Ramirez-Wong, Marius Ramirez, Regina Sanchez-Leija, Adriana Rugerio, R. Araceli Mauricio-Sanchez, Martin A. Hernandez-Landaverde, Arturo Carranza, John A. Pojman, Josue D. Mota-Morales, Gabriel Luna-Barcenas

Abstract:

Chitin films were prepared using alpha-chitin from shrimp shells as raw material and a simple method of precipitation-evaporation. Choline chloride: urea Deep Eutectic Solvent (DES) was used to disperse chitin and compared against hexafluoroisopropanol (HFIP). A careful analysis of the chemical and crystalline structure was followed along the synthesis of the films, revealing crystalline-phase transitions. The full conversion of alpha- to beta-, or alpha- to gamma-chitin structure were detected by XRD and NMR on the films. The synthesis of highly crystalline monophasic gamma-chitin films was achieved using a DES; whereas HFIP helps to promote the beta-phase. These results are encouraging to continue in the study of DES as good processing media to control the final properties of chitin based materials.

Keywords: chitin, deep eutectic solvent, polymorph, phase transformation

Procedia PDF Downloads 506

Authors: Jingting Luo, Yong Wang, Ying Wang

Abstract:

Pedestrian detection in image or video data is a very important and challenging task in security surveillance. The difficulty of this task is to locate and detect pedestrians of different scales in complex scenes accurately. To solve these problems, a deep neural network (RT-YOLOv3) is proposed to realize real-time pedestrian detection at different scales in security monitoring. RT-YOLOv3 improves the traditional YOLOv3 algorithm. Firstly, the deep residual network is added to extract vehicle features. Then six convolutional neural networks with different scales are designed and fused with the corresponding scale feature maps in the residual network to form the final feature pyramid to perform pedestrian detection tasks. This method can better characterize pedestrians. In order to further improve the accuracy and generalization ability of the model, a hybrid pedestrian data set training method is used to extract pedestrian data from the VOC data set and train with the INRIA pedestrian data set. Experiments show that the proposed RT-YOLOv3 method achieves 93.57% accuracy of mAP (mean average precision) and 46.52f/s (number of frames per second). In terms of accuracy, RT-YOLOv3 performs better than Fast R-CNN, Faster R-CNN, YOLO, SSD, YOLOv2, and YOLOv3. This method reduces the missed detection rate and false detection rate, improves the positioning accuracy, and meets the requirements of real-time detection of pedestrian objects.

Keywords: pedestrian detection, feature detection, convolutional neural network, real-time detection, YOLOv3

Procedia PDF Downloads 113

Authors: João Costa, Francisco Albuquerque, Ricardo J. Santos, Madalena M. Dias, José Carlos B. Lopes, Marcelo Costa

Abstract:

Nowadays, it is well recognized that carbon dioxide emissions, together with other greenhouse gases, are responsible for the dramatic climate changes that have been occurring over the past decades. Gas hydrates are currently seen as a promising and disruptive set of materials that can be used as a basis for developing new technologies for CO₂ capture and storage. Its potential as a clean and safe pathway for CCS is tremendous since it requires only water and gas to be mixed under favorable temperatures and mild high pressures. However, the hydrates formation process is highly exothermic; it releases about 2 MJ per kilogram of CO₂, and it only occurs in a narrow window of operational temperatures (0 - 10 °C) and pressures (15 to 40 bar). Efficient continuous hydrate production at a specific temperature range necessitates high heat transfer rates in mixing processes. Past technologies often struggled to meet this requirement, resulting in low productivity or extended mixing/contact times due to inadequate heat transfer rates, which consistently posed a limitation. Consequently, there is a need for more effective continuous hydrate production technologies in industrial applications. In this work, a network mixing continuous production technology has been shown to be viable for producing CO₂ hydrates. The structured mixer used throughout this work consists of a network of unit cells comprising mixing chambers interconnected by transport channels. These mixing features result in enhanced heat and mass transfer rates and high interfacial surface area. The mixer capacity emerges from the fact that, under proper hydrodynamic conditions, the flow inside the mixing chambers becomes fully chaotic and self-sustained oscillatory flow, inducing intense local laminar mixing. The device presents specific heat transfer rates ranging from 107 to 108 W⋅m⁻³⋅K⁻¹. A laboratory scale pilot installation was built using a device capable of continuously capturing 1 kg⋅h⁻¹ of CO₂, in an aqueous slurry of up to 20% in mass. The strong mixing intensity has proven to be sufficient to enhance dissolution and initiate hydrate crystallization without the need for external seeding mechanisms and to achieve, at the device outlet, conversions of 99% in CO₂. CO₂ dissolution experiments revealed that the overall liquid mass transfer coefficient is orders of magnitude larger than in similar devices with the same purpose, ranging from 1 000 to 12 000 h⁻¹. The present technology has shown itself to be capable of continuously producing CO₂ hydrates. Furthermore, the modular characteristics of the technology, where scalability is straightforward, underline the potential development of a modular hydrate-based CO₂ capture process for large-scale applications.

Keywords: network, mixing, hydrates, continuous process, carbon dioxide

Procedia PDF Downloads 19

Authors: Jonathan Sohn

Abstract:

Deep neural networks (DNNs) have shown state-of-the-art performance for many applications, including computer vision, natural language processing, and speech recognition. Recently, adversarial attacks have been studied in the context of deep neural networks, which aim to alter the results of deep neural networks by modifying the inputs slightly. For example, an adversarial attack on a DNN used for object detection can cause the DNN to miss certain objects. As a result, the reliability of DNNs is undermined by their lack of robustness against adversarial attacks, raising concerns about their use in safety-critical applications such as autonomous driving. In this paper, we focus on studying the adversarial attacks and defenses on DNNs for image classification. There are two types of adversarial attacks studied which are fast gradient sign method (FGSM) attack and projected gradient descent (PGD) attack. A DNN forms decision boundaries that separate the input images into different categories. The adversarial attack slightly alters the image to move over the decision boundary, causing the DNN to misclassify the image. FGSM attack obtains the gradient with respect to the image and updates the image once based on the gradients to cross the decision boundary. PGD attack, instead of taking one big step, repeatedly modifies the input image with multiple small steps. There is also another type of attack called the target attack. This adversarial attack is designed to make the machine classify an image to a class chosen by the attacker. We can defend against adversarial attacks by incorporating adversarial examples in training. Specifically, instead of training the neural network with clean examples, we can explicitly let the neural network learn from the adversarial examples. In our experiments, the digit recognition accuracy on the MNIST dataset drops from 97.81% to 39.50% and 34.01% when the DNN is attacked by FGSM and PGD attacks, respectively. If we utilize FGSM training as a defense method, the classification accuracy greatly improves from 39.50% to 92.31% for FGSM attacks and from 34.01% to 75.63% for PGD attacks. To further improve the classification accuracy under adversarial attacks, we can also use a stronger PGD training method. PGD training improves the accuracy by 2.7% under FGSM attacks and 18.4% under PGD attacks over FGSM training. It is worth mentioning that both FGSM and PGD training do not affect the accuracy of clean images. In summary, we find that PGD attacks can greatly degrade the performance of DNNs, and PGD training is a very effective way to defend against such attacks. PGD attacks and defence are overall significantly more effective than FGSM methods.

Keywords: deep neural network, adversarial attack, adversarial defense, adversarial machine learning

Procedia PDF Downloads 158

Authors: Hui Zhang, Sherif Beskhyroun

Abstract:

Structural Health monitoring systems (SHM) are commonly used to identify and assess structural damage. In terms of damage detection, SHM needs to periodically collect data from sensors placed in the structure as damage-sensitive features. This includes abnormal changes caused by the strain field and abnormal symptoms of the structure, such as damage and deterioration. Currently, deploying sensors on a large scale in a building structure is a challenge. In this study, a highly stretchable strain sensors are used in this study to collect data sets of strain generated on the surface of full-size reinforced concrete (RC) frames under extreme cyclic load application. This sensing sheet can be switched freely between the test bending strain and the axial strain to achieve two different configurations. On this basis, the deep neural network prediction model of the frame beam and frame column is established. The training results show that the method can accurately predict the strain value and has good generalization ability. The two deep neural network prediction models will also be deployed in the SHM system in the future as part of the intelligent strain sensor system.

Keywords: strain sensing sheets, deep neural networks, strain measurement, SHM system, RC frames

Procedia PDF Downloads 61

Authors: Pawan S. Nagda, Purnank S. Bhatt, Mit K. Shah

Abstract:

Earing defect in drawing process is highly undesirable not only because it adds on an additional trimming operation but also because the uneven material flow demands extra care. The objective of this work is to study the earing problem in the Deep Drawing of circular cup and to optimize the blank shape to reduce the earing. A finite element model is developed for 3-D numerical simulation of cup forming process in ABAQUS. Extra-deep-drawing (EDD) steel sheet has been used for simulation. Properties and tool design parameters were used as input for simulation. Earing was observed in the simulated cup and it was measured at various angles with respect to rolling direction. To reduce the earing defect initial blank shape was modified with the help of anisotropy coefficient. Modified blanks showed notable reduction in earing.

Keywords: anisotropy, deep drawing, earing, finite element simulation

Procedia PDF Downloads 353

Authors: Javad Saeidaskari, Mohammad Hassan Baziar

Abstract:

Soil structures have been frequently damaged during piping, earthquake and other types of failures. As far as adverse circumstances were developed subsequent to piping or other similar failure types, hydraulic parameters of soil such as hydraulic conductivity should be considered. As a result, acquiring an approach to diminish soil permeability is inevitable. There are many ground improvement methods to reduce seepage, which are classified under soil treatment and stabilization methods. Recently, one of the soil improvement methods is known as nanogeotechnology. This study aims to investigate the influence of Cloisite 30B nanoclay on permeability of compacted clayey sand soils. The samples are prepared by mixing two soil types, including Kaolin clay and Firouzkooh sand, in 1:9 and 1:5 clay:sand (by mass) proportions. In experimental procedure, initially, the optimum water content and maximum dry unit weight of each samples were obtained for compaction. Then, series of permeability tests were conducted by triaxial apparatus on prepared specimens with identical relative density of 95% of maximum dry density and water content of 1% wet of optimum for different weight percentages of nanoclay (1% to 4%). Therefore, in this paper, the effect of time on treated specimen was appraised, as well as two approaches of manual mixing and ball milling were compared to reveal the importance of dispersion issue. The results show that adding nanoclay up to 3%, as its optimum content, causes notable reduction in permeability (1.60e-03 to 5.51e-05 cm/s and 3.32e-04 to 8.44e-07 cm/s in samples with 1:9 and 1:5 mixture proportions, respectively). The hydraulic conductivity of treated clayey sand (1:5 mixture proportion with 3% nanoclay) decreases gradually from 8.44e-07 to 3.00e-07 cm/s within 90 days and then tends to be consistent. The influence of mixing method on permeability results shows that the utilization of ball mill mixing effectively leads to lower values than those of manual mixing, in other words, by adding 3% nanoclay, hydraulic conductivity of specimen declines from 8.44e-07 to 2.00e-07 cm/s. In order to evaluate the interaction between soil particles and, to ensure proper dispersion of nanoparticles through clayey sand mixture, they were magnified by means of scanning electron microscope (SEM). In conclusion, the nanoclay particles in vicinity of moisture can cause soil stabilization to prevent water penetration, which eventually result in lower usage of clay and operation costs.

Keywords: nanoclay, cloisite 30b, clayey sand, hydraulic conductivity

Procedia PDF Downloads 321

Authors: Sarwat un Nisa

Abstract:

One of the remarkable gifts that a human being is blessed with is the ability to speak using a combination of sounds. Different combinations of sounds combine to form a word which in turn make a sentence and therefore give birth to a language. A person can either be a monolingual, i.e., can speak one language or bilingual, i.e., can speak more than one language. Whether a person speaks one language or multiple languages or in whatever language a person speaks, the main aim is to communicate, express ideas, feelings or thoughts. Sometimes the choice of a language is deliberate and sometimes it is a habitual act. The language which is used to put our ideas across speaks many things about our cultural, linguistic and ethnic identities. It can never be claimed that bilinguals are better than monolinguals in terms of linguistic skills, bilinguals or multilinguals have more than one language at their disposal. Therefore, how effectively two languages are used by the same person keeps linguists always intrigued. The most prominent and common features found in the speech of bilingual speakers are code switching and code mixing. The aim of the present paper is to explore these features among the adolescent speakers of Kashmir. The reason for studying the linguistics behavior of adolescents is the age when a person is neither an adult nor a child. They want to drift away from the norms and make a new norm for themselves. Therefore, how their linguistics skills are influenced by their age is of great interest because it can set the trend for the future generation. Kashmir is a multilingual society where three languages, i.e., Kashmiri, Urdu, and English are regularly used by the speakers, especially the educated ones. Kashmiri is widely used at home or mostly among adults. Urdu is the official language, and English is used in schools and for most of the written official correspondences. Thus, it is not uncommon to find these three languages coming in contact with each other quite frequently. The language contact results in the code switching and code mixing. In this paper different aspects of code switching and code mixing are discussed. Research Method: The data were collected from the different districts of Kashmir. The informants did not have prior knowledge of the survey. The situation was spontaneous and natural. The topics were introduced by the interviewer to the group of informants which comprised of three participants. They were asked to discuss the topic, most of the times without any intervention of the interviewer. Along with conversations, the informants also filled in written questionnaires comprising sociolinguistic questions. Questionnaires were analysed to get an idea about the sociolinguistic attitude of the informants. Percentage, frequency, and average were used as statistical tools to analyse the data. Conclusions were drawn taking into consideration of interpretations of both speech samples and questionnaires.

Keywords: code mixing, code switching, Kashmir, bilingualism

Procedia PDF Downloads 112

Authors: Alexander Nitsche, Piotr-Robert Lazik, Harald Garrecht

Abstract:

The viscosity of the concrete has a great influence on the properties of the fresh concrete. Fresh concretes with low viscosity have a good flowability, whereas high viscosity has a lower flowability. Clearly, viscosity is directly linked to other parameters such as consistency, compaction, and workability of the concrete. The above parameters also depend very much on the energy induced during the mixing process and, of course, on the installation of the mixer itself. The University of Stuttgart has decided to investigate the influence of different mixing systems on the viscosity of various types of concrete, such as road concrete, self-compacting concrete, and lightweight concrete, using a rheometer and other testing methods. Each type is tested with three different mixers, and the rheological properties, namely consistency, and viscosity are determined. The aim of the study is to show that different types of concrete mixed with different types of mixers reach completely different yield points. Therefore, a 3 step procedure will be introduced. At first, various types of concrete mixtures and their differences are introduced. Then, the chosen suspension mixer and conventional mixers, which are going to be used in this paper, will be discussed. Lastly, the influence of the mixing system on the rheological properties of each of the select mix designs, as well as on fresh concrete, in general, will be presented.

Keywords: rheological properties, flowability, suspension mixer, viscosity

Procedia PDF Downloads 114

Authors: Shaik Ayesha Fathima, Shaik Noor Jahan, Duvvada Rajeswara Rao

Abstract:

Earth's environment and its evolution can be seen through satellite images in near real-time. Through satellite imagery, remote sensing data provide crucial information that can be used for a variety of applications, including image fusion, change detection, land cover classification, agriculture, mining, disaster mitigation, and monitoring climate change. The objective of this project is to propose a method for classifying satellite images according to multiple predefined land cover classes. The proposed approach involves collecting data in image format. The data is then pre-processed using data pre-processing techniques. The processed data is fed into the proposed algorithm and the obtained result is analyzed. Some of the algorithms used in satellite imagery classification are U-Net, Random Forest, Deep Labv3, CNN, ANN, Resnet etc. In this project, we are using the DeepLabv3 (Atrous convolution) algorithm for land cover classification. The dataset used is the deep globe land cover classification dataset. DeepLabv3 is a semantic segmentation system that uses atrous convolution to capture multi-scale context by adopting multiple atrous rates in cascade or in parallel to determine the scale of segments.

Keywords: area calculation, atrous convolution, deep globe land cover classification, deepLabv3, land cover classification, resnet 50

Procedia PDF Downloads 114

Authors: Okoro Ikechukwu Ralph, Fuat Kara

Abstract:

The discoveries of hydrocarbon reserves has clearly drifted offshore, and in deeper waters - areas where the industry still has limited knowledge; and that were hitherto, regarded as being out of reach. This shift presents significant and increased challenges in technology requirements needed to guarantee safety of personnel, environment and equipment; ensure high reliability of installed equipment; and provide high level of confidence in security of investment and company reputation. Nowhere are these challenges more apparent than on subsea well integrity and production performance. The past two decades has witnessed enormous rise in deep and ultra-deep water offshore field developments for the recovery of hydrocarbons. Subsea installed equipment at the seabed has been the technology of choice for these developments. This paper discusses the role of Subsea Control module (SCM) as a vital factor for deep-water well integrity and production performance. A case study for Deep-water well integrity and production performance is analysed.

Keywords: offshore reliability, production performance, subsea control module, well integrity

Procedia PDF Downloads 480

Authors: H. Lee, J. Park

Abstract:

In this present study, for the first time, we retrieved the Formaldehyde (HCHO) Vertical Column Density (HCHOVCD) using Pandora instruments in Seoul, a megacity in northeast Asia, for the period between 2012 and 2014 and investigated the temporal characteristics of HCHOVCD. HCHO Slant Column Density (HCHOSCD) was obtained using the Differential Optical Absorption Spectroscopy (DOAS) method. HCHOSCD was converted to HCHOVCD using geometric Air Mass Factor (AMFG) as Pandora is the direct-sun measurement. The HCHOVCDs is low at 12:00 Local Time (LT) and is high in the morning (10:00 LT) and late afternoon (16:00 LT) except for winter. The maximum (minimum) values of Pandora HCHOVCD are 2.68×1016 (1.63×10¹⁶), 3.19×10¹⁶ (2.23×10¹⁶), 2.00×10¹⁶ (1.26×10¹⁶), and 1.63×10¹⁶ (0.82×10¹⁶) molecules cm⁻² in spring, summer, autumn, and winter, respectively. In terms of seasonal variations, HCHOVCD was high in summer and low in winter which implies that photo-oxidation plays an important role in HCHO production in Seoul. In comparison with the Ozone Monitoring Instrument (OMI) measurements, the HCHOVCDs from the OMI are lower than those from Pandora. The correlation coefficient (R) between monthly HCHOVCDs values from Pandora and OMI is 0.61, with slop of 0.35. Furthermore, to understand HCHO mixing ratio within Planetary Boundary Layer (PBL) in Seoul, we converted Pandora HCHOVCDs to HCHO mixing ratio in the PBL using several meteorological input data from the Atmospheric InfraRed Sounder (AIRS). Seasonal HCHO mixing ratio in PBL converted from Pandora (OMI) HCHOVCDs are estimated to be 6.57 (5.17), 7.08 (6.68), 7.60 (4.70), and 5.00 (4.76) ppbv in spring, summer, autumn, and winter, respectively.

Keywords: formaldehyde, OMI, Pandora, remote sensing

Procedia PDF Downloads 125

Authors: Christian Neunzig, Simon Fahle, Jürgen Schulz, Matthias Möller, Bernd Kuhlenkötter

Abstract:

The increasing use of deep learning applications in production is becoming a competitive advantage. Predictive quality enables the assurance of product quality by using data-driven forecasts via machine learning models as a basis for decisions on test results. The use of real Bosch production data along the value chain of hydraulic valves is a promising approach to classifying the leakage of directional valves.

Keywords: artificial neural networks, classification, hydraulics, predictive quality, deep learning

Procedia PDF Downloads 200

Authors: Joy Cao, Min Zhou

Abstract:

Purpose: Acute Type A aortic dissection is a well-known cause of extremely high mortality rate. A highly accurate and cost-effective non-invasive predictor is critically needed so that the patient can be treated at earlier stage. Although various CFD approaches have been tried to establish some prediction frameworks, they are sensitive to uncertainty in both image segmentation and boundary conditions. Tedious pre-processing and demanding calibration procedures requirement further compound the issue, thus hampering their clinical applicability. Using the latest physics informed deep learning methods to establish an accurate and cost-effective predictor framework are amongst the main goals for a better Type A aortic dissection treatment. Methods: Via training a novel physics-informed deep residual network, with non-invasive 4D MRI displacement vectors as inputs, the trained model can cost-effectively calculate all these biomarkers: aortic blood pressure, WSS, and OSI, which are used to predict potential type A aortic dissection to avoid the high mortality events down the road. Results: The proposed deep learning method has been successfully trained and tested with both synthetic 3D aneurysm dataset and a clinical dataset in the aortic dissection context using Google colab environment. In both cases, the model has generated aortic blood pressure, WSS, and OSI results matching the expected patient’s health status. Conclusion: The proposed novel physics-informed deep residual network shows great potential to create a cost-effective, non-invasive predictor framework. Additional physics-based de-noising algorithm will be added to make the model more robust to clinical data noises. Further studies will be conducted in collaboration with big institutions such as Cleveland Clinic with more clinical samples to further improve the model’s clinical applicability.

Keywords: type-a aortic dissection, deep residual networks, blood flow modeling, data-driven modeling, non-invasive diagnostics, deep learning, artificial intelligence.

Procedia PDF Downloads 56

Authors: Thibaud Berthomier, Ali Mansour, Luc Bressollette, Frédéric Le Roy, Dominique Mottier, Léo Fréchier, Barthélémy Hermenault

Abstract:

Deep Venous Thrombosis (DVT) occurs when a thrombus is formed within a deep vein (most often in the legs). This disease can be deadly if a part or the whole thrombus reaches the lung and causes a Pulmonary Embolism (PE). This disorder, often asymptomatic, has multifactorial causes: immobilization, surgery, pregnancy, age, cancers, and genetic variations. Our project aims to relate the thrombus epidemiology (origins, patient predispositions, PE) to its structure using ultrasound images. Ultrasonography and elastography were collected using Toshiba Aplio 500 at Brest Hospital. This manuscript compares two classification approaches: spectral clustering and scattering operator. The former is based on the graph and matrix theories while the latter cascades wavelet convolutions with nonlinear modulus and averaging operators.

Keywords: deep venous thrombosis, ultrasonography, elastography, scattering operator, wavelet, spectral clustering

Procedia PDF Downloads 450

Authors: Vladimir Viktorov, Readul Mahmud, Carmen Visconte

Abstract:

In the present study, a new passive micromixer based on SAR principle, combining the operation concepts of known Chain and H mixers, called C-H micromixer, is developed and studied. The efficiency and the pressure drop of the C-H mixer along with two known SAR passive mixers named Chain and Tear-drop were investigated numerically at Reynolds numbers up to 100, taking into account species transport. At the same time experimental tests of the Chain and Tear-drop mixers were carried out at low Reynolds number, in the 0.1≤Re≤4.2 range. Numerical and experimental results coincide considerably, which validate the numerical simulation approach. Results show that mixing efficiency of the Tear-drop mixer is good except at the middle range of Reynolds number but pressure drop is too high; conversely the Chain mixer has moderate pressure drop but relatively low mixing efficiency at low and middle Re numbers. Whereas, the C-H mixer gives excellent mixing efficiency at all range of Re numbers. In addition, the C-H mixer shows respectively about 3 and 2 times lower pressure drop than the Tear-drop mixer and the Chain mixer.

Keywords: CFD, micromixing, passive micromixer, SAR

Procedia PDF Downloads 277

Authors: Mst. Tuhin Akter, Sharun Akter Khushbu, S. M. Shaqib

Abstract:

A real-time system architecture is highly effective for monitoring and detecting various damaged products or fruits that may deteriorate over time or become infected with diseases. Deep learning models have proven to be effective in building such architectures. However, building a deep learning model from scratch is a time-consuming and costly process. A more efficient solution is to utilize deep neural network (DNN) based transfer learning models in the real-time monitoring architecture. This study focuses on using a novel strawberry dataset to develop effective transfer learning models for the proposed real-time monitoring system architecture, specifically for evaluating and detecting strawberry freshness. Several state-of-the-art transfer learning models were employed, and the best performing model was found to be Xception, demonstrating higher performance across evaluation metrics such as accuracy, recall, precision, and F1-score.

Keywords: strawberry freshness evaluation, deep neural network, transfer learning, image augmentation

Procedia PDF Downloads 55

Authors: Sait Ozsoy, Asude Gokmen, Mehtap Yondem, Hanife A. Alkan, Gulnaz T. Javan

Abstract:

Deep neck infection often develops due to upper respiratory tract and odontogenic infections. Gastrointestinal System perforation can occur for many reasons and is in need of the early diagnosis and prompt surgical treatment. In both cases late or incorrect diagnosis may lead to increase morbidity and high mortality. A patient with a diagnosis of deep neck abscess died while under treatment due to sepsis and multiple organ failure. Autopsy finding showed duodenal ulcer and this is reported in the literature.

Keywords: peptic ulcer perforation, peritonitis, retropharyngeal abscess, sepsis

Procedia PDF Downloads 467

Authors: Reza Khoshnood, Zahra Khoshnood, Ali Hajinajaf, Farzad Fahim, Behdokht Hajinajaf, Farhad Fahim

Abstract:

In 2013, 24 fish samples were taken from two fishery regions in Bandar-Abbas and Bandar-Lengeh, the fishing grounds north of Hormoz Strait (Persian Gulf) near the Iranian coastline. The two flat fishes were oriental sole (Euryglossa orientalis) and deep flounder (Psettodes erumei). Using the ROPME method (MOOPAM) for chemical digestion, Cd concentration was measured with a nonflame atomic absorption spectrophotometry technique. The average concentration of Cd in the edible muscle tissue of deep flounder was measured in Bandar-Abbas and was found to be 0.15±.06 µg g-1. It was 0.1±.05 µg.g-1 in Bandar-Lengeh. The corresponding values for oriental sole were 0.2±0.13 and 0.13±0.11 µg.g-1. The average concentration of Cd in the liver tissue of deep flounder in Bandar-Abbas was 0.22±.05 µg g-1 and that in Bandar-Lengeh was 0.2±0.04 µg.g-1. The values for oriental sole were 0.31±0.09 and 0.24±0.13 µg g-1 in Bandar-Abbas and Bandar-Lengeh, respectively.

Keywords: trace metal, Euryglossa orientalis, Psettodes erumei, Persian Gulf

Procedia PDF Downloads 634

Authors: Shaik Jilani Basha, Inavolu Avinash, Alla Venu Sai Reddy, Bitragunta Taraka Ramu

Abstract:

We can use the neural style transfer technique to build a picture with the same "content" as the beginning image but the "style" of the picture we've chosen. Neural style transfer is a technique for merging the style of one image into another while retaining its original information. The only change is how the image is formatted to give it an additional artistic sense. The content image depicts the plan or drawing, as well as the colors of the drawing or paintings used to portray the style. It is a computer vision programme that learns and processes images through deep convolutional neural networks. To implement software, we used to train deep learning models with the train data, and whenever a user takes an image and a styled image, the output will be as the style gets transferred to the original image, and it will be shown as the output.

Keywords: neural networks, computer vision, deep learning, convolutional neural networks

Procedia PDF Downloads 53

Authors: Modupe O. Jimoh

Abstract:

During manhole overflows, its contents pollute the immediate environment. Understanding the pollutant transfer characteristics between manhole’s incoming sewer and the overflow is therefore of great importance. A square manhole with sides 388 mm by 388 mm and height 700 mm with an overflow facility was used in the laboratory to carry out overflow concentration measurements. Two scenarios were investigated using three flow rates. The first scenario corresponded to when the exit of the pipe becomes blocked and the only exit for the flow is the manhole. The second scenario is when there is an overflow in combination with a pipe exit. The temporal concentration measurements showed that the peak concentration of pollutants in the flow was attenuated between the inlet and the overflow. A deconvolution software was used to predict the Residence time distribution (RTD) and consequently the Cumulative Residence time distribution (CRTD). The CRTDs suggest that complete mixing is occurring between the pipe inlet and the overflow, like what is obtained in a low surcharged manhole. The results also suggest that an instantaneous stirred tank reactor model can describe the mixing characteristics.

Keywords: CRTDs, instantaneous stirred tank reactor model, overflow, square manholes, surcharge, temporal concentration profiles

Procedia PDF Downloads 109

Authors: M. A. Salman, G. Al-Nuwaibit, M. Safar, M. Rughaibi, A. Al-Mesri

Abstract:

Barium sulfate (BaSO₄) is a hard scaling usually precipitates on the surface of equipment in many industrial systems, as oil and gas production, desalination and cooling and boiler operation. It is a scale that extremely resistance to both chemical and mechanical cleaning. So, BaSO₄ is a problematic and expensive scaling. Although barium ions are present in most natural waters at a very low concentration as low as 0.008 mg/l, it could result of scaling problems in the presence of high concentration of sulfate ion or when mixing with incompatible waters as in oil produced water. The scaling potential of BaSO₄ using seawater at the intake of seven desalination plants in Kuwait, brine water and Kuwait oil produced water was calculated and compared then the best location in regards of barium sulfate scaling was reported. Finally, a physical treatment method (magnetic treatment method) and chemical treatment method were used to control BaSO₄ scaling using saturated solutions at different operating temperatures, flow velocities, feed pHs and different magnetic strengths. The results of the two methods were discussed, and the more economical one with the reasonable performance was recommended, which is the physical treatment method.

Keywords: magnetic field strength, flow velocity, retention time, barium sulfate

Procedia PDF Downloads 237

Authors: Kamal KC, Zhendong Yin, Dasen Li, Zhilu Wu

Abstract:

Intra-plant disease classification based on leaf images is a challenging computer vision task due to similarities in texture, color, and shape of leaves with a slight variation of leaf spot; and external environmental changes such as lighting and background noises. Deep convolutional neural network (DCNN) has proven to be an effective tool for binary classification. In this paper, two methods for binary classification of diseased plant leaves using DCNN are presented; model created from scratch and transfer learning. Our main contribution is a thorough evaluation of 4 networks created from scratch and transfer learning of 5 pre-trained models. Training and testing of these models were performed on a plant leaf images dataset belonging to 16 distinct classes, containing a total of 22,265 images from 8 different plants, consisting of a pair of healthy and diseased leaves. We introduce a deep CNN model, Optimized MobileNet. This model with depthwise separable CNN as a building block attained an average test accuracy of 99.77%. We also present a fine-tuning method by introducing the concept of a convolutional block, which is a collection of different deep neural layers. Fine-tuned models proved to be efficient in terms of accuracy and computational cost. Fine-tuned MobileNet achieved an average test accuracy of 99.89% on 8 pairs of [healthy, diseased] leaf ImageSet.

Keywords: deep convolution neural network, depthwise separable convolution, fine-grained classification, MobileNet, plant disease, transfer learning

Procedia PDF Downloads 157

Authors: Nurhak Altın, Ayhan Kucukmanisa, Oguzhan Urhan

Abstract:

It is important that highways are in good condition for traffic safety. Road crashes (road cracks, erosion of lane markings, etc.) can cause accidents by affecting driving. Image processing based methods for detecting road cracks are available in the literature. In this paper, a deep learning based road crack detection approach is proposed. YOLO (You Look Only Once) is adopted as core component of the road crack detection approach presented. The YOLO network structure, which is developed for object detection, is trained with road crack images as a new class that is not previously used in YOLO. The performance of the proposed method is compared using different training methods: using randomly generated weights and training their own pre-trained weights (transfer learning). A similar training approach is applied to the simplified version of the YOLO network model (tiny yolo) and the results of the performance are examined. The developed system is able to process 8 fps on NVIDIA Jetson TX1 development kit.

Keywords: deep learning, embedded platform, real-time processing, road crack detection

Procedia PDF Downloads 314

Authors: Susan Diamond

Abstract:

Machine Learning workloads have traditionally been run in high-performance computing (HPC) environments, where users log in to dedicated machines and utilize the attached GPUs to run training jobs on huge datasets. Training of large neural network models is very resource intensive, and even after exploiting parallelism and accelerators such as GPUs, a single training job can still take days. Consequently, the cost of hardware is a barrier to entry. Even when upfront cost is not a concern, the lead time to set up such an HPC environment takes months from acquiring hardware to set up the hardware with the right set of firmware, software installed and configured. Furthermore, scalability is hard to achieve in a rigid traditional lab environment. Therefore, it is slow to react to the dynamic change in the artificial intelligent industry. Watson Deep Learning as a service, a cloud-based deep learning platform that mitigates the long lead time and high upfront investment in hardware. It enables robust and scalable sharing of resources among the teams in an organization. It is designed for on-demand cloud environments. Providing a similar user experience in a multi-tenant cloud environment comes with its own unique challenges regarding fault tolerance, performance, and security. Watson Deep Learning as a service tackles these challenges and present a deep learning stack for the cloud environments in a secure, scalable and fault-tolerant manner. It supports a wide range of deep-learning frameworks such as Tensorflow, PyTorch, Caffe, Torch, Theano, and MXNet etc. These frameworks reduce the effort and skillset required to design, train, and use deep learning models. Deep Learning as a service is used at IBM by AI researchers in areas including machine translation, computer vision, and healthcare. 

Keywords: deep learning, machine learning, cognitive computing, model training

Procedia PDF Downloads 180

Authors: Ryoma Tokonami, Teruya Goto, Tatsuhiro Takahashi,

Abstract:

For enhancing the mechanical property ofcarbon fiber reinforced plastic (CFRP), the surface modification of carbon fiber (CF) by multi-walled carbon nanotube (MWCNT) has received considerable attention using direct MWCNT growth on CF with a catalysis, MWCNT electrophoresis, and layer-by-layer of MWCNT with reactive polymers, etc. Among above approaches, the layer-by-layer method is the simplest process, however, the amount of MWCNTs on CF is very little, resulting in the small amount of improvement of the mechanical property of the composite. The remaining amount of MWCNT on CF after melt mixing of CF (short fiber) with thermoplastic matrix polymer was not examined clearly in the former studies. The present research aims to propose an effective layer-by-layer chemical grafting of a highly reactive oxazoline polymer, which has not been used before, and MWCNTs onto CF using the highly reactivity of oxazoline and COOH on the surface of CF and MWCNTs.With layer-by-layer method, the first uniform chemically bonded mono molecular layer on carbon fiber was formed by chemical surface reaction of carbon fiber, a reactive oxazoline polymer solution between COOH of carbon fiber and oxazoline. The second chemically bonded uniform layer of MWCNTs on the first layer was prepared through the first layer coated carbon fiber in MWCNT dispersion solution by chemical reaction between oxazoline and COOH of MWCNTs. The quantitative analysis of MWCNTs on carbon fiber was performed, showing 0.44 wt.% of MWCNTs based on carbon fiber, which is much larger amount compared with the former studies in layer-by-layer method. In addition, MWCNTs were also observed uniform coating on carbon fiber by scanning electron micrograph (SEM). Carbon fiber composites were prepared by melting mixing using polystyrene (PS) as a thermoplastic matrix because of easy removal of PS by solvent for additional analysis, resulting the 20% of enhancement of tensile strength and modulus by tensile strength test. It was confirmed bySEM the layer-by-layer structure on carbon fibers were remained after the melt mixing by removing PS with a solvent. As a conclusion, the effectiveness for the enhancement of the mechanical properties of CF(short fiber)/PS composite using the highly reactive oxazoline polymer for the first layer and MWCNT for the second layer, which act as the physical anchor, was demonstrated.

Keywords: interface, layer-by-layer, multi walled carbon nanotubes (MWCNTs), oxazoline

Procedia PDF Downloads 165

Authors: Barham J. Nareeman, Ilham I. Mohammed

Abstract:

Excavations in urban developed area are generally supported by deep excavation walls such as; diaphragm wall, bored piles, soldier piles and sheet piles. In some cases, these walls may be braced by internal braces or tie back anchors. Tie back anchors are by far the predominant method for wall support, the large working space inside the excavation provided by a tieback anchor system has a significant construction advantage. This paper aims to analyze a deep excavation bracing system of contiguous pile wall braced by pre-stressed tie back anchors, which is a part of a huge residential building project, located in Turkey/Gaziantep province. The contiguous pile wall will be constructed with a length of 270 m that consists of 285 piles, each having a diameter of 80 cm, and a center to center spacing of 95 cm. The deformation analysis was carried out by a finite element analysis tool using PLAXIS. In the analysis, beam element method together with an elastic perfect plastic soil model and Soil Hardening Model was used to design the contiguous pile wall, the tieback anchor system, and the soil. The two soil clusters which are limestone and a filled soil were modelled with both Hardening soil and Mohr Coulomb models. According to the basic design, both soil clusters are modelled as drained condition. The simulation results show that the maximum horizontal movement of the walls and the maximum settlement of the ground are convenient with 300 individual case histories which are ranging between 1.2mm and 2.3mm for walls, and 15mm and 6.5mm for the settlements. It was concluded that tied-back contiguous pile wall can be satisfactorily modelled using Hardening soil model.

Keywords: deep excavation, finite element, pre-stressed tie back anchors, contiguous pile wall, PLAXIS, horizontal deflection, ground settlement

Procedia PDF Downloads 222

Authors: Lianzhong Zhang, Chao Huang

Abstract:

Sea-land segmentation is a basic step in many tasks such as sea surface monitoring and ship detection. The existing sea-land segmentation algorithms have poor segmentation accuracy, and the parameter adjustments are cumbersome and difficult to meet actual needs. Also, the current sea-land segmentation adopts traditional deep learning models that use Convolutional Neural Networks (CNN). At present, the transformer architecture has achieved great success in the field of natural images, but its application in the field of radar images is less studied. Therefore, this paper proposes a sea-land segmentation method based on the transformer architecture to strengthen edge supervision. It uses a self-attention mechanism with a gating strategy to better learn relative position bias. Meanwhile, an additional edge supervision branch is introduced. The decoder stage allows the feature information of the two branches to interact, thereby improving the edge precision of the sea-land segmentation. Based on the Gaofen-3 satellite image dataset, the experimental results show that the method proposed in this paper can effectively improve the accuracy of sea-land segmentation, especially the accuracy of sea-land edges. The mean IoU (Intersection over Union), edge precision, overall precision, and F1 scores respectively reach 96.36%, 84.54%, 99.74%, and 98.05%, which are superior to those of the mainstream segmentation models and have high practical application values.

Keywords: SAR, sea-land segmentation, deep learning, transformer

Procedia PDF Downloads 136