Search results for: convolutional architectures

Authors: Joseph George, Anne Kotteswara Roa

Abstract:

Skin disease is one of the most common and popular kinds of health issues faced by people nowadays. Skin cancer (SC) is one among them, and its detection relies on the skin biopsy outputs and the expertise of the doctors, but it consumes more time and some inaccurate results. At the early stage, skin cancer detection is a challenging task, and it easily spreads to the whole body and leads to an increase in the mortality rate. Skin cancer is curable when it is detected at an early stage. In order to classify correct and accurate skin cancer, the critical task is skin cancer identification and classification, and it is more based on the cancer disease features such as shape, size, color, symmetry and etc. More similar characteristics are present in many skin diseases; hence it makes it a challenging issue to select important features from a skin cancer dataset images. Hence, the skin cancer diagnostic accuracy is improved by requiring an automated skin cancer detection and classification framework; thereby, the human expert’s scarcity is handled. Recently, the deep learning techniques like Convolutional neural network (CNN), Deep belief neural network (DBN), Artificial neural network (ANN), Recurrent neural network (RNN), and Long and short term memory (LSTM) have been widely used for the identification and classification of skin cancers. This survey reviews different DL techniques for skin cancer identification and classification. The performance metrics such as precision, recall, accuracy, sensitivity, specificity, and F-measures are used to evaluate the effectiveness of SC identification using DL techniques. By using these DL techniques, the classification accuracy increases along with the mitigation of computational complexities and time consumption.

Keywords: skin cancer, deep learning, performance measures, accuracy, datasets

Procedia PDF Downloads 98

Authors: Kenan Morani, Esra Kaya Ayana

Abstract:

This study aimed to develop a novel pipeline for COVID-19 detection using a large and rigorously annotated database of computed tomography (CT) images. The pipeline consists of UNet-based segmentation, lung extraction, and a classification part, with the addition of optional slice removal techniques following the segmentation part. In this work, a batch normalization was added to the original UNet model to produce lighter and better localization, which is then utilized to build a full pipeline for COVID-19 diagnosis. To evaluate the effectiveness of the proposed pipeline, various segmentation methods were compared in terms of their performance and complexity. The proposed segmentation method with batch normalization outperformed traditional methods and other alternatives, resulting in a higher dice score on a publicly available dataset. Moreover, at the slice level, the proposed pipeline demonstrated high validation accuracy, indicating the efficiency of predicting 2D slices. At the patient level, the full approach exhibited higher validation accuracy and macro F1 score compared to other alternatives, surpassing the baseline. The classification component of the proposed pipeline utilizes a convolutional neural network (CNN) to make final diagnosis decisions. The COV19-CT-DB dataset, which contains a large number of CT scans with various types of slices and rigorously annotated for COVID-19 detection, was utilized for classification. The proposed pipeline outperformed many other alternatives on the dataset.

Keywords: classification, computed tomography, lung extraction, macro F1 score, UNet segmentation

Procedia PDF Downloads 103

Authors: Ilgoun Kim, Jongpil Jeong

Abstract:

Recently, a connected industrial parks (CIPs) architecture using new technologies such as RFID, cloud computing, CPS, Big Data, 5G 5G, IIOT, VR-AR, and ventral AI algorithms based on IoT has been proposed. This researcher noted the vehicle junction problem (VJP) as a more specific detail of the CIPs architectural models. The VJP noted by this researcher includes 'efficient AI physical connection challenges for vehicles' through ventilation, 'financial and financial issues with complex vehicle physical connections,' and 'welfare and working conditions of the performing personnel involved in complex vehicle physical connections.' In this paper, we propose a public solution architecture for the 'electronic financial problem of complex vehicle physical connections' as a detailed task during the vehicle junction problem (VJP). The researcher sought solutions to businesses, consumers, and Korean social problems through technological advancement. We studied how the beneficiaries of technological development can benefit from technological development with many consumers in Korean society and many small and small Korean company managers, not some specific companies. In order to more specifically implement the connected industrial parks (CIPs) architecture using the new technology, we noted the vehicle junction problem (VJP) within the smart factory industrial complex and noted the process of achieving the vehicle junction problem performance among several electronic processes. This researcher proposes a more detailed, integrated public finance enterprise architecture among the overall CIPs architectures. The main details of the public integrated financial enterprise architecture were largely organized into four main categories: 'business', 'data', 'technique', and 'finance'.

Keywords: enterprise architecture, IT Finance, smart logistics, CIPs

Procedia PDF Downloads 137

Authors: Debanjana Chatterjee

Abstract:

Vernacular settlements in India often take the form that is dictated by the climate they are in. India, with its vast cultural diversity and various climatic regions, offers a wide range of vernacular architecture. This paper focuses on two main geographical regions: Kashmir and Kerala. They bring together myriad challenges of climatic and social characteristics to incorporate into their vernacular architectures, which are still relevant despite the advent of globalization and modernization. Scholars like William Wurster and Catherine Bauer even claimed that all the traditional buildings in these places have the kind of urbanity, which is dignified and elegant but also lively and human that every architect would like to achieve if they knew how. With modernization, and with a greater ease of construction, a reduction in labor, and the apparent robustness of contemporary construction techniques, people have, however, become increasingly tentative in respect of vernacular architecture. And yet modern architecture has typically led to energize intensive structures without much consideration to the location and surroundings of the structure itself. In contrary, Laurie Baker, the British-born Indian architect, had shown us the way to integrate the knowledge of vernacular when he developed his designs based on the traditional architecture of Kerala, respecting the local climate and environment. This paper also explores his technical creativity in his design of Center for Development Studies (CDS) in Trivandrum. Hence, in order to protect and conserve our rich cultural and architectural heritage, the elements of vernacular should be incorporated into the contemporary planning and architecture for sustainable building design. The provision should be made to incorporate vernacular architecture and traditional knowledge in the policies. Ultimately, the policymakers, planners, and architects should consider this incorporation of traditional vernacular and contemporary sustainability in their work for the betterment of society now.

Keywords: vernacular, architecture, sustainable development, Kashmir and Kerala, climate, Laurie Baker

Procedia PDF Downloads 150

Authors: Teja Sai Puligadda

Abstract:

Satellite images in detail can serve an important role in the geographic study. Quantitative and qualitative information provided by the satellite and remote sensing images minimizes the complexity of work and time. Data/images are captured at regular intervals by satellite remote sensing systems, and the amount of data collected is often enormous, and it expands rapidly as technology develops. Interpreting remote sensing images, geographic data mining, and researching distinct vegetation types such as agricultural and forests are all part of satellite image categorization. One of the biggest challenge data scientists faces while classifying satellite images is finding the best suitable classification algorithms based on the available that could able to classify images with utmost accuracy. In order to categorize satellite images, which is difficult due to the sheer volume of data, many academics are turning to deep learning machine algorithms. As, the CNN algorithm gives high accuracy in image recognition problems and automatically detects the important features without any human supervision and the ANN algorithm stores information on the entire network (Abhishek Gupta., 2020), these two deep learning algorithms have been used for satellite image classification. This project focuses on remote sensing through Deep Neural Networks i.e., ANN and CNN with Deep Sat (SAT-4) Airborne dataset for classifying images. Thus, in this project of classifying satellite images, the algorithms ANN and CNN are implemented, evaluated & compared and the performance is analyzed through evaluation metrics such as Accuracy and Loss. Additionally, the Neural Network algorithm which gives the lowest bias and lowest variance in solving multi-class satellite image classification is analyzed.

Keywords: artificial neural network, convolutional neural network, remote sensing, accuracy, loss

Procedia PDF Downloads 127

Authors: Hongmei Gu, Shaobo Liang, Richard Bergman

Abstract:

With current worldwide trend in designs to have net-zero emission buildings to mitigate climate change, widespread use of mass timber products, such as Cross Laminated Timber (CLT), or Nail Laminated Timber (NLT) or Dowel Laminated Timber (DLT) in buildings have been proposed as one approach in reducing Greenhouse Gas (GHG) emissions. Consequentially, mass timber building designs are being adopted more and more by architectures in North America, especially for mid- to high-rise buildings where concrete and steel buildings are currently prevalent, but traditional light-frame wood buildings are not. Wood buildings and their associated wood products have tended to have lower environmental impacts than competing energy-intensive materials. It is common practice to conduct life cycle assessments (LCAs) and life cycle cost analyses on buildings with traditional structural materials like concrete and steel in the building design process. Mass timber buildings with lower environmental impacts, especially GHG emissions, can contribute to the Net Zero-emission goal for the world-building sector. However, the economic impacts from CLT mass timber buildings still vary from the life-cycle cost perspective and environmental trade-offs associated with GHG emissions. This paper quantified the Total Life Cycle Cost and cradle-to-grave GHG emissions of a pre-designed CLT mass timber building and compared it to a functionally-equivalent concrete building. The Total life cycle Eco-cost-efficiency is defined in this study and calculated to discuss the trade-offs for the net-zero emission buildings in a holistic view for both environmental and economic impacts. Mass timber used in buildings for the United States is targeted to the materials from the nation’s sustainable managed forest in order to benefit both national and global environments and economies.

Keywords: GHG, economic impact, eco-cost-efficiency, total life-cycle costs

Procedia PDF Downloads 111

Authors: Gong Zhilin, Jing Yang, Jian Yin

Abstract:

The purpose of this paper is to develop a novel system for credit card fraud detection based on sequential modeling of data using hybrid deep learning models. The projected model encapsulates five major phases are pre-processing, imbalance-data handling, feature extraction, optimal feature selection, and fraud detection with an ensemble classifier. The collected raw data (input) is pre-processed to enhance the quality of the data through alleviation of the missing data, noisy data as well as null values. The pre-processed data are class imbalanced in nature, and therefore they are handled effectively with the K-means clustering-based SMOTE model. From the balanced class data, the most relevant features like improved Principal Component Analysis (PCA), statistical features (mean, median, standard deviation) and higher-order statistical features (skewness and kurtosis). Among the extracted features, the most optimal features are selected with the Self-improved Arithmetic Optimization Algorithm (SI-AOA). This SI-AOA model is the conceptual improvement of the standard Arithmetic Optimization Algorithm. The deep learning models like Long Short-Term Memory (LSTM), Convolutional Neural Network (CNN), and optimized Quantum Deep Neural Network (QDNN). The LSTM and CNN are trained with the extracted optimal features. The outcomes from LSTM and CNN will enter as input to optimized QDNN that provides the final detection outcome. Since the QDNN is the ultimate detector, its weight function is fine-tuned with the Self-improved Arithmetic Optimization Algorithm (SI-AOA).

Keywords: credit card, data mining, fraud detection, money transactions

Procedia PDF Downloads 104

Authors: Ethan James

Abstract:

Over one billion people worldwide suffer from some level of vision loss or blindness as a result of progressive retinal diseases. Many patients, particularly in developing areas, are incorrectly diagnosed or undiagnosed whatsoever due to unconventional diagnostic tools and screening methods. Artificial intelligence (AI) based on deep learning (DL) convolutional neural networks (CNN) have recently gained a high interest in ophthalmology for its computer-imaging diagnosis, disease prognosis, and risk assessment. Optical coherence tomography (OCT) is a popular imaging technique used to capture high-resolution cross-sections of retinas. In ophthalmology, DL has been applied to fundus photographs, optical coherence tomography, and visual fields, achieving robust classification performance in the detection of various retinal diseases including macular degeneration, diabetic retinopathy, and retinitis pigmentosa. However, there is no complete diagnostic model to analyze these retinal images that provide a diagnostic accuracy above 90%. Thus, the purpose of this project was to develop an AI model that utilizes machine learning techniques to automatically diagnose specific retinal diseases from OCT scans. The algorithm consists of neural network architecture that was trained from a dataset of over 20,000 real-world OCT images to train the robust model to utilize residual neural networks with cyclic pooling. This DL model can ultimately aid ophthalmologists in diagnosing patients with these retinal diseases more quickly and more accurately, therefore facilitating earlier treatment, which results in improved post-treatment outcomes.

Keywords: artificial intelligence, deep learning, imaging, medical devices, ophthalmic devices, ophthalmology, retina

Procedia PDF Downloads 152

Authors: Tarik Sipahi

Abstract:

Photonics research continues to push the boundaries of optical science, and the development of metasurface technology has emerged as a transformative force in this domain. The work presents the intricacies of a unified metasurface design tailored for efficient polarization and color control in optical systems. The proposed unified metasurface serves as a singular, nanoengineered optical element capable of simultaneous polarization modulation and color encoding. Leveraging principles from metamaterials and nanophotonics, this design allows for unprecedented control over the behavior of light at the subwavelength scale. The metasurface's spatially varying architecture enables seamless manipulation of both polarization states and color wavelengths, paving the way for a paradigm shift in optical system design. The advantages of this unified metasurface are diverse and impactful. By consolidating functions that traditionally require multiple optical components, the design streamlines optical systems, reducing complexity and enhancing overall efficiency. This approach is particularly promising for applications where compactness, weight considerations, and multifunctionality are crucial. Furthermore, the proposed unified metasurface design not only enhances multifunctionality but also addresses key challenges in optical system design, offering a versatile solution for applications demanding compactness and lightweight structures. The metasurface's capability to simultaneously manipulate polarization and color opens new possibilities in diverse technological fields. The research contributes to the evolution of optical science by showcasing the transformative potential of metasurface technology, emphasizing its role in reshaping the landscape of optical system architectures. This work represents a significant step forward in the ongoing pursuit of pushing the boundaries of photonics, providing a foundation for future innovations in compact and efficient optical devices.

Keywords: metasurface, nanophotonics, optical system design, polarization control

Procedia PDF Downloads 26

Authors: Gurukiran Veerapur, Nytik Birudavolu, Seetharam U. N., Chandravva Hebbi, R. Praneeth Reddy

Abstract:

This work presents a robust approach for the recognition of characters in Telugu and Kannada, two South Indian scripts with structural similarities in characters. To recognize the characters exhaustive datasets are required, but there are only a few publicly available datasets. As a result, we decided to create a dataset for one language (source language),train the model with it, and then test it with the target language.Telugu is the target language in this work, whereas Kannada is the source language. The suggested method makes use of Canny edge features to increase character identification accuracy on pictures with noise and different lighting. A dataset of 45,150 images containing printed Kannada characters was created. The Nudi software was used to automatically generate printed Kannada characters with different writing styles and variations. Manual labelling was employed to ensure the accuracy of the character labels. The deep learning models like CNN (Convolutional Neural Network) and Visual Attention neural network (VAN) are used to experiment with the dataset. A Visual Attention neural network (VAN) architecture was adopted, incorporating additional channels for Canny edge features as the results obtained were good with this approach. The model's accuracy on the combined Telugu and Kannada test dataset was an outstanding 97.3%. Performance was better with Canny edge characteristics applied than with a model that solely used the original grayscale images. The accuracy of the model was found to be 80.11% for Telugu characters and 98.01% for Kannada words when it was tested with these languages. This model, which makes use of cutting-edge machine learning techniques, shows excellent accuracy when identifying and categorizing characters from these scripts.

Keywords: base characters, modifiers, guninthalu, aksharas, vattakshara, VAN

Procedia PDF Downloads 27

Authors: Simone Huber, Bianca Schnalzer, Baptiste Alcalde, Sten Hanke, Lampros Mpaltadoros, Thanos G. Stavropoulos, Spiros Nikolopoulos, Ioannis Kompatsiaris, Lina Pérez- Breva, Vallivana Rodrigo-Casares, Jaime Fons-Martínez, Jeroen de Bruin

Abstract:

Developing new medicine and health solutions and improving patient health currently rely on the successful execution of clinical trials, which generate relevant safety and efficacy data. For their success, recruitment and retention of participants are some of the most challenging aspects of protocol adherence. Main barriers include: i) lack of awareness of clinical trials; ii) long distance from the clinical site; iii) the burden on participants, including the duration and number of clinical visits and iv) high dropout rate. Most of these aspects could be addressed with a new paradigm, namely the Remote Decentralized Clinical Trials (RDCTs). Furthermore, the COVID-19 pandemic has highlighted additional advantages and challenges for RDCTs in practice, allowing participants to join trials from home and not depend on site visits, etc. Nevertheless, RDCTs should follow the process and the quality assurance of conventional clinical trials, which involve several processes. For each part of the trial, the Building Blocks, existing software and technologies were assessed through a systematic search. The technology needed to perform RDCTs is widely available and validated but is yet segmented and developed in silos, as different software solutions address different parts of the trial and at various levels. The current paper is analyzing the availability of technology to perform RDCTs, identifying gaps and providing an overview of Basic Building Blocks and functionalities that need to be covered to support the described processes.

Keywords: architectures and frameworks for health informatics systems, clinical trials, information and communications technology, remote decentralized clinical trials, technology availability

Procedia PDF Downloads 183

Authors: Svetlana Bialkova, Marnix van Gisbergen

Abstract:

Advanced Virtual Reality (VR) technologies offer compelling multisensory and interactive experiences applicable in various fields from education to entertainment. However, serious VR applications within the financial sector are scarce, and managing ‘real’ business services with(in) VR is a challenge inviting further investigation. The current research addresses this challenge, by exploring the key parameters influencing the VR business power and the development of appropriate VR applications in real financial business. We conducted profound investigation of both B2B and B2C needs, and how these could be met. In three studies, we have approached experts from leading international banks (finance to computer specialists), and their (potential) customers. Study 1 included focus group discussions with experts. First, participants could experience different VR devices such as Samsung Gear VR, then a structured discussion was held. The outcomes are analyzed and summarized in a portfolio. Study 2 further used the portfolio analyzer to profile the management of real business services with(in) VR. Again experts participated, where first being introduced with Samsung Gear, then experiencing it and being interviewed. Based on the outcomes, a survey was developed to interview (potential) customers and test ideas created (Study 3). The results suggest that developing proper system architectures to connect people and to connect devices is crucial for building up powerful business with(in) VR. From one side, connecting devices, e.g., pairing mobile Head Mounted Displays for VR with smart-phones and/or wearable technologies would be appropriate way “to have” customers anywhere, anytime with a brand and/or business. Developing VR Apps, providing detailed real time visualization of performance and infrastructure types could enable 3D VR navigation, 3D contents viewing, but also being opportunity for connecting people in collaborative platforms. The outcomes of the current research are summarized in a model which could be applied to unlock the real business power of VR.

Keywords: business power, B2B, B2C, VR applications

Procedia PDF Downloads 268

Authors: Sindhu Govardhan, Kenneth G. Crowther

Abstract:

Modernisation of the water sector is leading to increased connectivity and integration of emerging technologies with traditional ones, leading to new security risks. The convergence of Information Technology (IT) with Operation Technology (OT) results in solutions that are spread across larger geographic areas, increasingly consist of interconnected Industrial Internet of Things (IIOT) devices and software, rely on the integration of legacy with modern technologies, use of complex supply chain components leading to complex architectures and communication paths. The result is that multiple parties collectively own and operate these emergent technologies, threat actors find new paths to exploit, and traditional cybersecurity controls are inadequate. Our approach is to explicitly identify and draw data flows that cross trust boundaries between owners and operators of various aspects of these emerging and interconnected technologies. On these data flows, we layer potential attack vectors to create a frame of reference for evaluating possible risks against connected technologies. Finally, we identify where existing controls, mitigations, and other remediations exist across industry partners (e.g., suppliers, product vendors, integrators, water utilities, and regulators). From these, we are able to understand potential gaps in security, the roles in the supply chain that are most likely to effectively remediate those security gaps, and test cases to evaluate and strengthen security across these partners. This informs a “shared responsibility” solution that recognises that security is multi-layered and requires collaboration to be successful. This shared responsibility security framework improves visibility, understanding, and control across the entire supply chain, and particularly for those water utilities that are accountable for safe and continuous operations.

Keywords: cyber security, shared responsibility, IIOT, threat modelling

Procedia PDF Downloads 47

Authors: Osvaldo N. Oliveira Jr.

Abstract:

The control of molecular architecture inherent in some experimental methods to produce nanostructured films has had great impact on devices of various types, including sensors and biosensors. The self-assembly monolayers (SAMs) and the electrostatic layer-by-layer (LbL) techniques, for example, are now routinely used to produce tailored architectures for biosensing where biomolecules are immobilized with long-lasting preserved activity. Enzymes, antigens, antibodies, peptides and many other molecules serve as the molecular recognition elements for detecting an equally wide variety of analytes. The principles of detection are also varied, including electrochemical methods, fluorescence spectroscopy and impedance spectroscopy. In this presentation an overview will be provided of biosensors made with nanostructured films to detect antibodies associated with tropical diseases and HIV, in addition to detection of analytes of medical interest such as cholesterol and triglycerides. Because large amounts of data are generated in the biosensing experiments, use has been made of computational and statistical methods to optimize performance. Multidimensional projection techniques such as Sammon´s mapping have been shown more efficient than traditional multivariate statistical analysis in identifying small concentrations of anti-HIV antibodies and for distinguishing between blood serum samples of animals infected with two tropical diseases, namely Chagas´ disease and Leishmaniasis. Optimization of biosensing may include a combination of another information visualization method, the Parallel Coordinate technique, with artificial intelligence methods in order to identify the most suitable frequencies for reaching higher sensitivity using impedance spectroscopy. Also discussed will be the possible convergence of technologies, through which machine learning and other computational methods may be used to treat data from biosensors within an expert system for clinical diagnosis.

Keywords: clinical diagnosis, information visualization, nanostructured films, layer-by-layer technique

Procedia PDF Downloads 306

Authors: Esmeralda Hysenbelliu

Abstract:

The aim of this paper is to present the QoE (Quality of Experience) IPTV SDN-based media streaming server enhanced architecture for configuring, controlling, management and provisioning the improved delivery of IPTV service application with low cost, low bandwidth, and high security. Furthermore, it is given a virtual QoE IPTV SDN-based topology to provide an improved IPTV service based on QoE Control and Management of multimedia services functionalities. Inside OpenFlow SDN Controller there are enabled in high flexibility and efficiency Service Load-Balancing Systems; based on the Loading-Balance module and based on GeoIP Service. This two Load-balancing system improve IPTV end-users Quality of Experience (QoE) with optimal management of resources greatly. Through the key functionalities of OpenFlow SDN controller, this approach produced several important features, opportunities for overcoming the critical QoE metrics for IPTV Service like achieving incredible Fast Zapping time (Channel Switching time) < 0.1 seconds. This approach enabled Easy and Powerful Transcoding system via FFMPEG encoder. It has the ability to customize streaming dimensions bitrates, latency management and maximum transfer rates ensuring delivering of IPTV streaming services (Audio and Video) in high flexibility, low bandwidth and required performance. This QoE IPTV SDN-based media streaming architecture unlike other architectures provides the possibility of Channel Exchanging between several IPTV service providers all over the word. This new functionality brings many benefits as increasing the number of TV channels received by end –users with low cost, decreasing stream failure time (Channel Failure time < 0.1 seconds) and improving the quality of streaming services.

Keywords: improved quality of experience (QoE), OpenFlow SDN controller, IPTV service application, softwarization

Procedia PDF Downloads 121

Authors: Laxmi R. Kasaraneni

Abstract:

The Internet connects all the networks, including the nation’s critical infrastructure that are used extensively by not only a nation’s government and military to protect sensitive information and execute missions, but also the primary infrastructure that provides services that enable modern conveniences such as education, potable water, electricity, natural gas, and financial transactions. It has become the central nervous system for the government, the citizens, and the industries. When it is attacked, the effects can ripple far and wide impacts not only to citizens’ well-being but nation’s economy, civil infrastructure, and national security. As such, these critical services may be targeted by malicious hackers during cyber warfare, it is imperative to not only protect them and mitigate any immediate or potential threats, but to also understand the current or potential impacts beyond the IT networks or the organization. The Nation’s IT infrastructure which is now vital for communication, commerce, and control of our physical infrastructure, is highly vulnerable to attack. While existing technologies can address some vulnerabilities, fundamentally new architectures and technologies are needed to address the larger structural insecurities of an infrastructure developed in a more trusting time when mass cyber attacks were not foreseen. This research is intended to improve the core functions of the Internet and critical-sector information systems by providing a clear path to create a safe, secure, and resilient cyber environment that help stakeholders at all levels of government, and the private sector work together to develop the cybersecurity capabilities that are key to our economy, national security, and public health and safety. This research paper also emphasizes the present and future cyber security threats, the capabilities and goals of cyber attackers, a strategic concept and steps to implement cybersecurity for maximum effectiveness, enabling technologies, some strategic assumptions and critical challenges, and the future of cyberspace.

Keywords: critical challenges, critical infrastructure, cyber security, enabling technologies, national security

Procedia PDF Downloads 268

Authors: Hongyun Li, Zhibin Jiang

Abstract:

The travel of passengers in the urban rail transit network is influenced by changes in network structure and operational status, and the response of individual travel preferences to these changes also varies. Firstly, the influence of the suspension of urban rail transit line sections on passenger travel along the line is analyzed. Secondly, passenger travel trajectories containing multi-dimensional semantics are described based on network UD data. Next, passenger panel data based on spatio-temporal sequences is constructed to achieve frequent passenger clustering. Then, the Graph Convolutional Network (GCN) is used to model and identify the changes in travel modes of different types of frequent passengers. Finally, taking Shanghai Metro Line 11 as an example, the travel behavior patterns of frequent passengers after the Huaqiao section shutdown during the COVID-19 epidemic are analyzed. The results showed that after the section shutdown, most passengers would transfer to the nearest Anting station for boarding, while some passengers would transfer to other stations for boarding or cancel their travels directly. Among the passengers who transferred to Anting station for boarding, most of passengers maintained the original normalized travel mode, a small number of passengers waited for a few days before transferring to Anting station for boarding, and only a few number of passengers stopped traveling at Anting station or transferred to other stations after a few days of boarding on Anting station. The results can provide a basis for understanding urban rail transit passenger travel patterns and improving the accuracy of passenger flow prediction in abnormal operation scenarios.

Keywords: urban rail transit, section shutdown, frequent passenger, travel behavior pattern

Procedia PDF Downloads 53

Authors: Chun-Ming Huang, Yi-Jun Liu, Yi-Jie Hsieh, Jin-Ju Chue, Wei-Lin Lai, Chun-Yu Chen, Chih-Chyau Yang, Chien-Ming Wu

Abstract:

The market of wearable devices has been growing rapidly in two years. The integration of sensors and wearable devices has become the trend of the next technology products. Thus, the academics and industries are eager to cultivate talented persons in sensing technology. Currently, academic and industries have more and more demands on the integrations of versatile sensors and applications, especially for the teams who focus on the development of sensor circuit architectures. These teams tape-out many MEMs sensors chips through the chip fabrication service from National Chip Implementation Center (CIC). However, most of these teams are only able to focus on the circuit design of MEMs sensors; they lack the key support of further system demonstration. This paper follows the CIC’s main mission of promoting the chip/system advanced design technology and aims to establish the environments of the modularized sensing system platform and the system design flow with the measurement and calibration technology. These developed environments are used to support these research teams and help academically advanced sensor designs to perform the system demonstration. Thus, the research groups can promote and transfer their advanced sensor designs to industrial and further derive the industrial economic values. In this paper, the modularized sensing platform is proposed to enable the system demonstration for advanced sensor chip design. The environment of sensor measurement and calibration is established for academic to achieve an accurate sensor result. Two reference sensor designs cooperated with the modularized sensing platform are given to show the sensing system integration and demonstration. These developed environments and platforms are currently provided to academics in Taiwan, and so that the academics can obtain a better environment to perform the system demonstration and improve the research and teaching quality.

Keywords: modularized sensing platform, sensor design and calibration, sensor system, sensor system design flow

Procedia PDF Downloads 211

Authors: Yehjune Heo

Abstract:

As biometric systems become widely deployed, the security of identification systems can be easily attacked by various spoof materials. This paper contributes to finding a reliable and practical anti-spoofing method using Convolutional Neural Networks (CNNs) based on the types of loss functions and optimizers. The types of CNNs used in this paper include AlexNet, VGGNet, and ResNet. By using various loss functions including Cross-Entropy, Center Loss, Cosine Proximity, and Hinge Loss, and various loss optimizers which include Adam, SGD, RMSProp, Adadelta, Adagrad, and Nadam, we obtained significant performance changes. We realize that choosing the correct loss function for each model is crucial since different loss functions lead to different errors on the same evaluation. By using a subset of the Livdet 2017 database, we validate our approach to compare the generalization power. It is important to note that we use a subset of LiveDet and the database is the same across all training and testing for each model. This way, we can compare the performance, in terms of generalization, for the unseen data across all different models. The best CNN (AlexNet) with the appropriate loss function and optimizers result in more than 3% of performance gain over the other CNN models with the default loss function and optimizer. In addition to the highest generalization performance, this paper also contains the models with high accuracy associated with parameters and mean average error rates to find the model that consumes the least memory and computation time for training and testing. Although AlexNet has less complexity over other CNN models, it is proven to be very efficient. For practical anti-spoofing systems, the deployed version should use a small amount of memory and should run very fast with high anti-spoofing performance. For our deployed version on smartphones, additional processing steps, such as quantization and pruning algorithms, have been applied in our final model.

Keywords: anti-spoofing, CNN, fingerprint recognition, loss function, optimizer

Procedia PDF Downloads 109

Authors: Attef Kouadria, Yehya Bouteghrine, Amar Manaa, Tarek Mouats, Djalel Eddine Tria, Hamid Abdelhafid Ghouti

Abstract:

Personal protection systems have been used in several forms for centuries. The need for light-weight composite structures has been in great demand due to their weight and high mechanical properties ratios in comparison to heavy and cumbersome steel plates. In this regard, lighter ceramic plates with a backing plate made of high strength polymeric fibers, mostly aramids, are widely used for protection against ballistic threats. This study aims to improve the ballistic performance of ceramic/composite plates subjected to ballistic impact by reducing the back face deformation (BFD) measured after each test. A new hybridization technique was developed in this investigation to increase the energy absorption capabilities of the backing plates. The hybridization consists of combining different types of aramid fabrics with different linear densities of aramid fibers (Dtex) and areal densities with an epoxy resin to form the backing plate. Therefore, several composite structures architectures were prepared and tested. For better understanding the effect of the hybridization, a serial of tensile, compression, and shear tests were conducted to determine the mechanical properties of the homogeneous composite materials prepared from different fabrics. It was found that the hybridization allows the backing plate to combine between the mechanical properties of the used fabrics. Aramid fabrics with higher Dtex were found to increase the mechanical strength of the backing plate, while those with lower Dtex found to enhance the lateral wave dispersion ratio due to their lower areal density. Therefore, the back face deformation was significantly reduced in comparison to a homogeneous composite plate.

Keywords: aramid fabric, ballistic impact, back face deformation, body armor, composite, mechanical testing

Procedia PDF Downloads 120

Authors: Hakim Herbane

Abstract:

Efforts to achieve sustainability in architecture must prove their effectiveness despite various methods attempted. Biomimicry, which looks to successful natural models to promote sustainability and innovation, faces obstacles in implementing sustainability despite its restorative approach to the relationship between humans and nature. In Nunavik, Inuit communities are exploring a sustainable production system that aligns with their aspirations and meets their demands of human, technological, technical, economic, and ecological factors. Biomimicry holds promise in line with Inuit philosophy, but its failure to implement sustainability requires further investigations to remedy its deficiencies. Our literature review underscores the importance of involving the community in defining sustainability and determining the best methods for its implementation. Additionally, vernacular architecture shows valuable orientations for achieving sustainability. Moreover, reintegrating Inuit communities and their traditional architectural practices, which have successfully balanced their built environment's diverse needs and constraints, could pave the way for a sustainable Inuit-built environment in Nunavik and advance architectural biomimicry principles simultaneously. This research aims at establishing a sustainability monitoring tool for Nordic architectural process by analyzing Inuit vernacular and biomimetic architecture, in addition to the input of stakeholders involved in Inuit architecture production in Nunavik, especially Inuit. The goal is to create a practical tool (an index) to aid in designing sustainable architecture, taking into account environmental, social, and economic perspectives. Furthermore, the study seeks to authenticate strong, sustainable design principles of vernacular and biomimetic architectures. The literature review uncovered challenges and identified new opportunities. The forthcoming discourse will focus on the careful and considerate incorporation of Inuit communities’ perceptions and indigenous building practices into our methodology and the latest findings of our research.

Keywords: sustainability, biomimicry, vernacular architecture, community involvement

Procedia PDF Downloads 27

Authors: Pengfei Meng, Shuangcheng Jia, Qian Li

Abstract:

We proposed a real-time high-precision semantic segmentation network based on a multi-resolution feature fusion module, the auxiliary feature extracting module, upsampling module, and atrous spatial pyramid pooling (ASPP) module. We designed a feature fusion structure, which is integrated with sufficient features of different resolutions. We also studied the effect of side-branch structure on the network and made discoveries. Based on the discoveries about the side-branch of the network structure, we used a side-branch auxiliary feature extraction layer in the network to improve the effectiveness of the network. We also designed upsampling module, which has better results than the original upsampling module. In addition, we also re-considered the locations and number of atrous spatial pyramid pooling (ASPP) modules and modified the network structure according to the experimental results to further improve the effectiveness of the network. The network presented in this paper takes the backbone network of Bisenetv2 as a basic network, based on which we constructed a network structure on which we made improvements. We named this network deep multiple-resolution bilateral networks for real-time, referred to as DMBR-Net. After experimental testing, our proposed DMBR-Net network achieved 81.2% mIoU at 119FPS on the Cityscapes validation dataset, 80.7% mIoU at 109FPS on the CamVid test dataset, 29.9% mIoU at 78FPS on the COCOStuff test dataset. Compared with all lightweight real-time semantic segmentation networks, our network achieves the highest accuracy at an appropriate speed.

Keywords: multi-resolution feature fusion, atrous convolutional, bilateral networks, pyramid pooling

Procedia PDF Downloads 111

Authors: Lori W. Gordon, Karen A. Jones

Abstract:

Today’s high-speed data connectivity depends on a vast global network of infrastructure across space, air, land, and sea, with undersea cable infrastructure (UCI) serving as the primary means for intercontinental and ‘long-haul’ communications. The UCI landscape is changing and includes an increasing variety of state actors, such as the growing economies of Brazil, Russia, India, China, and South Africa. Non-state commercial actors, such as hyper-scale content providers including Google, Facebook, Microsoft, and Amazon, are also seeking to control their data and networks through significant investments in submarine cables. Active investments by both state and non-state actors will invariably influence the growth, geopolitics, and security of this sector. Beyond these hyper-scale content providers, there are new commercial satellite communication providers. These new players include traditional geosynchronous (GEO) satellites that offer broad coverage, high throughput GEO satellites offering high capacity with spot beam technology, low earth orbit (LEO) ‘mega constellations’ – global broadband services. And potential new entrants such as High Altitude Platforms (HAPS) offer low latency connectivity, LEO constellations offer high-speed optical mesh networks, i.e., ‘fiber in the sky.’ This paper focuses on understanding the role of submarine cables within the larger context of the global data commons, spanning space, terrestrial, air, and sea networks, including an analysis of national security policy and geopolitical implications. As network operators and commercial and government stakeholders plan for emerging technologies and architectures, hedging risks for future connectivity will ensure that our data backbone will be secure for years to come.

Keywords: communications, global, infrastructure, technology

Procedia PDF Downloads 55

Authors: Animish Jain

Abstract:

This study deals with using Multiplicative Weight Update within artificial intelligence and machine learning to create models that can diagnose skin cancer using microscopic images of cancer samples. In this study, the multiplicative weight update method is used to take the predictions of multiple models to try and acquire more accurate results. Logistic Regression, Convolutional Neural Network (CNN), and Support Vector Machine Classifier (SVMC) models are employed within the Multiplicative Weight Update system. These models are trained on pictures of skin cancer from the ISIC-Archive, to look for patterns to label unseen scans as either benign or malignant. These models are utilized in a multiplicative weight update algorithm which takes into account the precision and accuracy of each model through each successive guess to apply weights to their guess. These guesses and weights are then analyzed together to try and obtain the correct predictions. The research hypothesis for this study stated that there would be a significant difference in the accuracy of the three models and the Multiplicative Weight Update system. The SVMC model had an accuracy of 77.88%. The CNN model had an accuracy of 85.30%. The Logistic Regression model had an accuracy of 79.09%. Using Multiplicative Weight Update, the algorithm received an accuracy of 72.27%. The final conclusion that was drawn was that there was a significant difference in the accuracy of the three models and the Multiplicative Weight Update system. The conclusion was made that using a CNN model would be the best option for this problem rather than a Multiplicative Weight Update system. This is due to the possibility that Multiplicative Weight Update is not effective in a binary setting where there are only two possible classifications. In a categorical setting with multiple classes and groupings, a Multiplicative Weight Update system might become more proficient as it takes into account the strengths of multiple different models to classify images into multiple categories rather than only two categories, as shown in this study. This experimentation and computer science project can help to create better algorithms and models for the future of artificial intelligence in the medical imaging field.

Keywords: artificial intelligence, machine learning, multiplicative weight update, skin cancer

Procedia PDF Downloads 49

Authors: Getaneh Berie Tarekegn

Abstract:

A precise localization system is crucial for many artificial intelligence Internet of Things (AI-IoT) applications in the era of smart cities. Their applications include traffic monitoring, emergency alarming, environmental monitoring, location-based advertising, intelligent transportation, and smart health care. The most common method for providing continuous positioning services in outdoor environments is by using a global navigation satellite system (GNSS). Due to nonline-of-sight, multipath, and weather conditions, GNSS systems do not perform well in dense urban, urban, and suburban areas.This paper proposes a generative AI-based positioning scheme for large-scale wireless settings using fingerprinting techniques. In this article, we presented a semi-supervised deep convolutional generative adversarial network (S-DCGAN)-based radio map construction method for real-time device localization. It also employed a reliable signal fingerprint feature extraction method with t-distributed stochastic neighbor embedding (t-SNE), which extracts dominant features while eliminating noise from hybrid WLAN and long-term evolution (LTE) fingerprints. The proposed scheme reduced the workload of site surveying required to build the fingerprint database by up to 78.5% and significantly improved positioning accuracy. The results show that the average positioning error of GAILoc is less than 0.39 m, and more than 90% of the errors are less than 0.82 m. According to numerical results, SRCLoc improves positioning performance and reduces radio map construction costs significantly compared to traditional methods.

Keywords: location-aware services, feature extraction technique, generative adversarial network, long short-term memory, support vector machine

Procedia PDF Downloads 23

Authors: Ertuğ Yıldırım, Dile Kara, Salih Zeki Yıldız

Abstract:

Metal containing polymers (MCPs) are macro molecules usually containing metal-ligand coordination units and are a multidisciplinary research field mainly based at the interface between coordination chemistry and polymer science. The progress of this area has also been reinforced by the growth of several other closely related disciplines including macro molecular engineering, crystal engineering, organic synthesis, supra molecular chemistry and colloidal and material science. Schiff base ligands are very effective in constructing supra molecular architectures such as coordination polymers, double helical and triple helical complexes. In addition, Schiff base derivatives incorporating a fluorescent moiety are appealing tools for optical sensing of metal ions. MCPs are well-known systems in which the combinations of local parameters are possible by means of fluoro metric techniques. Generally, without incorporation of the fluorescent groups with polymers is unspecific, and it is not useful to analyze their fluorescent properties. Therefore, it is necessary to prepare a new type epoxy polymers with fluorescent groups in terms of metal sensing prop and the other photo chemical applications. In the present study metal containing polymers were prepared via poly functional monomeric Schiff base metal chelate complexes in the presence of dis functional monomers such as diglycidyl ether Bisphenol A (DGEBA). The synthesized complexes and polymers were characterized by FTIR, UV-VIS and mass spectroscopies. The preparations of epoxy polymers have been carried out at 185 °C. The prepared composites having sharp and narrow excitation/emission properties are expected to be applicable in various systems such as heat-resistant polymers and photo voltaic devices. The prepared composite is also ideal for various applications, easily prepared, safe, and maintain good fluorescence properties.

Keywords: Schiff base ligands, crystal engineering, fluorescence properties, Metal Containing Polymers (MCPs)

Procedia PDF Downloads 324

Authors: Yu-Cheng Chen, Tzu-Ping Lin, Feng-Yi Lin, Chih-Yu Chen

Abstract:

Due to the urban heat island effect caused by highly development of city, the heat stress increased in recent year rapidly. Resulting in a sharp raise of the energy used in urban area. The heat stress during summer time exacerbated the usage of air conditioning and electric equipment, which caused more energy consumption and anthropogenic heat. Therefore, an accurate and simple method to measure energy used in urban area can be helpful for the architectures and urban planners to develop better energy efficiency goals. This research applies the combination of airborne LiDAR data and thermal imager to provide an innovate method to estimate energy consumption. Owing to the high resolution of remote sensing data, the accurate current volume and total floor area and the surface temperature of building derived from LiDAR and thermal imager can be herein obtained to predict energy used. In the estimate process, the LiDAR data will be divided into four type of land cover which including building, road, vegetation, and other obstacles. In this study, the points belong to building were selected to overlay with the land use information; therefore, the energy consumption can be estimated precisely with the real value of total floor area and energy use index for different use of building. After validating with the real energy used data from the government, the result shows the higher building in high development area like commercial district will present in higher energy consumption, caused by the large quantity of total floor area and more anthropogenic heat. Furthermore, because of the surface temperature can be warm up by electric equipment used, this study also applies the thermal image of building to find the hot spots of energy used and make the estimation method more complete.

Keywords: urban heat island, urban planning, LiDAR, thermal imager, energy consumption

Procedia PDF Downloads 219

Authors: Efe Çukur

Abstract:

This study aims to discuss and compare i) widespread and generic design, particularly industrial design education in relation to the specific needs of the automotive/transportation industry, and ii) an automotive/transportation design education model within and under to provide the conditions of design education and automotive industry, especially in Turkey and T.R.N.C. The automotive industry is the 11th largest in the world ($1.51 trillion). One of the most important departments in this industry, along with sales, marketing and engineering, is the design department. The automotive industry is known as the locomotive industry, but there is a non-automotive design department on the academic side of Turkey. This suggestion; includes the presentation of a program proposal that meets the needs of the industry for Turkey and T.R.N.C., the second largest automobile manufacturing country in Europe. On the education side, industrial design education has become a generic title. Automotive design studios are divided into several subgroups. Even in the higher graduate education, the automotive design departments get their subgroups like exterior design and interior design. Transportation design, which is a subfield of industrial design, is offered as higher education in transportation design departments, particularly in America and Europe. In these departments, the curriculum is shaped to the needs of the sectors. Higher education transportation design programs began in the mid-20th century. Until those high education programs...Until these high education programs, the industry has adapted architectures and engineers for designer workloads. Still today transportation design graduates are not the majority of the design studios. The content of the study is an in-depth comparison of these institutions and how the requirements, demands of the industry are met in this regard and revealed. Some of the institutions are selected from Europe and US. To be analyzed under the headings of staff, courses, syllabus, University-Industry collaboration, and location selection. The study includes short, mid, and long term proposals and a hypothesis for discussion. In short, the study will not only provide a wide comparative scope of information on generic and specialized aspects of design education in different countries but also propose a higher education model for automotive / transportation design with solid data of requirements, methodology, and structure regarding learning outcomes, and especially industry cooperation.

Keywords: design education, automotive - transportation design programs, transportation design, automotive industry in Turkey /T.R.N.C., automotive design education in Turkey /T.R.N.C.

Procedia PDF Downloads 69

Authors: Gregor Kosec

Abstract:

Simple and robust numerical approach for solving flow problems is presented, where involved physical fields are represented through the local approximation functions, i.e., the considered field is approximated over a local support domain. The approximation functions are then used to evaluate the partial differential operators. The type of approximation, the size of support domain, and the type and number of basis function can be general. The solution procedure is formulated completely through local computational operations. Besides local numerical method also the pressure velocity is performed locally with retaining the correct temporal transient. The complete locality of the introduced numerical scheme has several beneficial effects. One of the most attractive is the simplicity since it could be understood as a generalized Finite Differences Method, however, much more powerful. Presented methodology offers many possibilities for treating challenging cases, e.g. nodal adaptivity to address regions with sharp discontinuities or p-adaptivity to treat obscure anomalies in physical field. The stability versus computation complexity and accuracy can be regulated by changing number of support nodes, etc. All these features can be controlled on the fly during the simulation. The presented methodology is relatively simple to understand and implement, which makes it potentially powerful tool for engineering simulations. Besides simplicity and straightforward implementation, there are many opportunities to fully exploit modern computer architectures through different parallel computing strategies. The performance of the method is presented on the lid driven cavity problem, backward facing step problem, de Vahl Davis natural convection test, extended also to low Prandtl fluid and Darcy porous flow. Results are presented in terms of velocity profiles, convergence plots, and stability analyses. Results of all cases are also compared against published data.

Keywords: fluid flow, meshless, low Pr problem, natural convection

Procedia PDF Downloads 211

Authors: Malik Muhammad Arslan, Muneeb Ullah, Dai Shihan, Daniyal Haider, Xiaodong Yang

Abstract:

Chest X-rays are instrumental in the detection and monitoring of a wide array of diseases, including viral infections such as COVID-19, tuberculosis, pneumonia, lung cancer, and various cardiac and pulmonary conditions. To enhance the accuracy of diagnosis, artificial intelligence (AI) algorithms, particularly deep learning models like Convolutional Neural Networks (CNNs), are employed. However, these deep learning models demand a substantial and varied dataset to attain optimal precision. Generative Adversarial Networks (GANs) can be employed to create new data, thereby supplementing the existing dataset and enhancing the accuracy of deep learning models. Nevertheless, GANs have their limitations, such as issues related to stability, convergence, and the ability to distinguish between authentic and fabricated data. In order to overcome these challenges and advance the detection and classification of CXR normal and abnormal images, this study introduces a distinctive technique known as DCWGAN (Diverse Conditional Wasserstein GAN) for generating synthetic chest X-ray (CXR) images. The study evaluates the effectiveness of this Idiosyncratic DCWGAN technique using the ResNet50 model and compares its results with those obtained using the traditional GAN approach. The findings reveal that the ResNet50 model trained on the DCWGAN-generated dataset outperformed the model trained on the classic GAN-generated dataset. Specifically, the ResNet50 model utilizing DCWGAN synthetic images achieved impressive performance metrics with an accuracy of 0.961, precision of 0.955, recall of 0.970, and F1-Measure of 0.963. These results indicate the promising potential for the early detection of diseases in CXR images using this Inimitable approach.

Keywords: CNN, classification, deep learning, GAN, Resnet50

Procedia PDF Downloads 50