Search results for: deep networks

Authors: Shi-Uk Lee, Chae Young Lim

Abstract:

Objective: The aim of this study was to evaluate the changes of lumbar deep muscle thickness and cross-sectional area using ultrasonography with magnetic stimulation. Methods: To evaluate the changes of lumbar deep muscle by using magnetic stimulation, 12 healthy volunteers (39.6±10.0 yrs) without low back pain during 3 months participated in this study. All the participants were checked with X-ray and electrophysiologic study to confirm that they had no problems with their back. Magnetic stimulation was done on the L5 and S1 root with figure-eight coil as previous study. To confirm the proper motor root stimulation, the surface electrode was put on the tibialis anterior (L5) and abductor hallucis muscles (S1) and the hot spots of magnetic stimulation were found with 50% of maximal magnetic stimulation and determined the stimulation threshold lowering the magnetic intensity by 5%. Ultrasonography was used to assess the changes of L5 and S1 lumbar multifidus (superficial and deep) cross-sectional area and thickness with maximal magnetic stimulation. Cross-sectional area (CSA) and thickness was evaluated with image acquisition program, ImageJ software (National Institute of Healthy, USA). Wilcoxon signed-rank was used to compare outcomes between before and after stimulations. Results: The mean minimal threshold was 29.6±3.8% of maximal stimulation intensity. With minimal magnetic stimulation, thickness of L5 and S1 deep multifidus (DM) were increased from 1.25±0.20, 1.42±0.23 cm to 1.40±0.27, 1.56±0.34 cm, respectively (P=0.005, P=0.003). CSA of L5 and S1 DM were also increased from 2.26±0.18, 1.40±0.26 cm2 to 2.37±0.18, 1.56±0.34 cm2, respectively (P=0.002, P=0.002). However, thickness of L5 and S1 superficial multifidus (SM) were not changed from 1.92±0.21, 2.04±0.20 cm to 1.91±0.33, 1.96±0.33 cm (P=0.211, P=0.199) and CSA of L5 and S1 were also not changed from 4.29±0.53, 5.48±0.32 cm2 to 4.42±0.42, 5.64±0.38 cm2. With maximal magnetic stimulation, thickness of L5, S1 of DM and SM were increased (L5 DM, 1.29±0.26, 1.46±0.27 cm, P=0.028; L5 SM, 2.01±0.42, 2.24±0.39 cm, P=0.005; S1 DM, 1.29±0.19, 1.67±0.29 P=0.002; S1 SM, 1.90±0.36, 2.30±0.36, P=0.002). CSA of L5, S1 of DM and SM were also increased (all P values were 0.002). Conclusions: Deep lumbar muscles could be stimulated with lumbar motor root magnetic stimulation. With minimal stimulation, thickness and CSA of lumbosacral deep multifidus were increased in this study. Further studies are needed to confirm whether the similar results in chronic low back pain patients are represented. Lumbar magnetic stimulation might have strengthening effect of deep lumbar muscles with no discomfort.

Keywords: magnetic stimulation, lumbar multifidus, strengthening, ultrasonography

Procedia PDF Downloads 371

Authors: Sam Khozama, Ali M. Mayya

Abstract:

Breast cancer is one of the most common types in women. Early prediction of breast cancer helps physicians detect cancer in its early stages. Big cancer data needs a very powerful tool to analyze and extract predictions. Machine learning and deep learning are two of the most efficient tools for predicting cancer based on textual data. In this study, we developed a fusion model of two machine learning and deep learning models. To obtain the final prediction, Long-Short Term Memory (LSTM) and ensemble learning with hyper parameters optimization are used, and score-level fusion is used. Experiments are done on the Breast Cancer Surveillance Consortium (BCSC) dataset after balancing and grouping the class categories. Five different training scenarios are used, and the tests show that the designed fusion model improved the performance by 3.3% compared to the individual models.

Keywords: machine learning, deep learning, cancer prediction, breast cancer, LSTM, fusion

Procedia PDF Downloads 164

Authors: Yan Bachtiar Muslih, Hangga Wijaya, Trio Fani, Putri Agustin

Abstract:

Demand of Energy in Indonesia always increases 5-6% a year, but production of conventional energy always decreases 3-5% a year, it means that conventional energy in 20-40 years ahead will not able to complete all energy demand in Indonesia, one of the solve way is using unconventional energy that is gas hydrate, gas hydrate is gas that form by biogenic process, gas hydrate stable in condition with extremely depth and low temperature, gas hydrate can form in two condition that is in pole condition and in deep-sea condition, wherein this research will focus in gas hydrate that association with methane form methane hydrate in deep-sea condition and usually form in depth between 150-2000 m, this research will focus in process of methane hydrate formation that is biogenic process and the important role of deep-sea sediment so can produce accumulation of methane hydrate, methane hydrate usually will be accumulated in find sediment in deep-sea environment with condition high-pressure and low-temperature this condition too usually make methane hydrate change into white nodule, methodology of this research is geology field work and laboratory analysis, from geology field work will get sample data consist of 10-15 samples from Kerek Formation outcrops as random for imagine the condition of deep-sea environment that influence the methane hydrate formation and also from geology field work will get data of measuring stratigraphy in outcrops Kerek Formation too from this data will help to imagine the process in deep-sea sediment like energy flow, supply sediment, and etc, and laboratory analysis is activity to analyze all data that get from geology field work, the result of this research can used to exploration activity of methane hydrate in another prospect deep-sea environment in Indonesia.

Keywords: methane hydrate, deep-sea sediment, kerek formation, sub-basin of kendeng, central java, Indonesia

Procedia PDF Downloads 463

Authors: Tanish Mittal, Preyansh Agrawal, Esha Pahwa, Aarya Makwana

Abstract:

Multi-Scale and U-shaped Networks are widely used in various image restoration problems, including deblurring. Keeping in mind the wide range of applications, we present a comparison of these architectures and their effects on image deblurring. We also introduce a new block called as NFResblock. It consists of a Fast Fourier Transformation layer and a series of modified Non-Linear Activation Free Blocks. Based on these architectures and additions, we introduce NFResnet and NFResnet+, which are modified multi-scale and U-Net architectures, respectively. We also use three differ-ent loss functions to train these architectures: Charbonnier Loss, Edge Loss, and Frequency Reconstruction Loss. Extensive experiments on the Deep Video Deblurring dataset, along with ablation studies for each component, have been presented in this paper. The proposed architectures achieve a considerable increase in Peak Signal to Noise (PSNR) ratio and Structural Similarity Index (SSIM) value.

Keywords: multi-scale, Unet, deblurring, FFT, resblock, NAF-block, nfresnet, charbonnier, edge, frequency reconstruction

Procedia PDF Downloads 138

Authors: Rongbo Shen, Jianhua Yao, Kezhou Yan, Kuan Tian, Cheng Jiang, Ke Zhou

Abstract:

This study proposes a generative adversarial networks (GAN) framework to perform synthetic sampling in feature space, i.e., feature augmentation, to address the class imbalance problem in medical image analysis. A feature extraction network is first trained to convert images into feature space. Then the GAN framework incorporates adversarial learning to train a feature generator for the minority class through playing a minimax game with a discriminator. The feature generator then generates features for minority class from arbitrary latent distributions to balance the data between the majority class and the minority class. Additionally, a data cleaning technique, i.e., Tomek link, is employed to clean up undesirable conflicting features introduced from the feature augmentation and thus establish well-defined class clusters for the training. The experiment section evaluates the proposed method on two medical image analysis tasks, i.e., mass classification on mammogram and cancer metastasis classification on histopathological images. Experimental results suggest that the proposed method obtains superior or comparable performance over the state-of-the-art counterparts. Compared to all counterparts, our proposed method improves more than 1.5 percentage of accuracy.

Keywords: class imbalance, synthetic sampling, feature augmentation, generative adversarial networks, data cleaning

Procedia PDF Downloads 128

Authors: Gorkem Algan, Ilkay Ulusoy, Saban Gonul, Banu Turgut, Berker Bakbak

Abstract:

Computer vision systems recently made a big leap thanks to deep neural networks. However, these systems require correctly labeled large datasets in order to be trained properly, which is very difficult to obtain for medical applications. Two main reasons for label noise in medical applications are the high complexity of the data and conflicting opinions of experts. Moreover, medical imaging datasets are commonly tiny, which makes each data very important in learning. As a result, if not handled properly, label noise significantly degrades the performance. Therefore, a label-noise-robust learning algorithm that makes use of the meta-learning paradigm is proposed in this article. The proposed solution is tested on retinopathy of prematurity (ROP) dataset with a very high label noise of 68%. Results show that the proposed algorithm significantly improves the classification algorithm's performance in the presence of noisy labels.

Keywords: deep learning, label noise, robust learning, meta-learning, retinopathy of prematurity

Procedia PDF Downloads 162

Authors: Cahit Perkgöz

Abstract:

Artificial intelligence has brought new approaches to solving problems in almost every research field in recent years. One of these topics is shape design and optimization, which has the possibility of applications in many fields, such as nanotechnology and electronics. A properly constructed cost function can eliminate the need for labeled data required in deep learning and create desired shapes. In this work, the network parameters are optimized differentially, which differs from traditional approaches. The methods are tested for physics-related structures and successful results are obtained. This work is supported by Eskişehir Technical University scientific research project (Project No: 20ADP090)

Keywords: deep learning, shape design, optimization, artificial intelligence

Procedia PDF Downloads 154

Authors: Naci Büyükkaracığan

Abstract:

Artificial neural networks (ANN) is an electrical model based on the human brain nervous system and working principle. Artificial neural networks have been the subject of an active field of research that has matured greatly over the past 55 years. ANN now is used in many fields. But, it has been viewed that artificial neural networks give better results in particular optimization and control systems. There are requirements of optimization and control system in many of the area forming the subject of civil engineering applications. In this study, the first artificial intelligence systems are widely used in the solution of civil engineering systems were examined with the basic principles and technical aspects. Finally, the literature reviews for applications in the field of civil engineering were conducted and also artificial intelligence techniques were informed about the study and its results.

Keywords: artificial neural networks, civil engineering, Fuzzy logic, statistics

Procedia PDF Downloads 415

Authors: Noureddine Mohtaram, Jeremy Patrix, Jerome Verny

Abstract:

As an enormous number of online services have been developed into web applications, security problems based on web applications are becoming more serious now. Most intrusion detection systems rely on each request to find the cyber-attack rather than on user behavior, and these systems can only protect web applications against known vulnerabilities rather than certain zero-day attacks. In order to detect new attacks, we analyze the HTTP protocols of web servers to divide them into two categories: normal attacks and malicious attacks. On the other hand, the quality of the results obtained by deep learning (DL) in various areas of big data has given an important motivation to apply it to cybersecurity. Deep learning for attack detection in cybersecurity has the potential to be a robust tool from small transformations to new attacks due to its capability to extract more high-level features. This research aims to take a new approach, deep learning to cybersecurity, to classify these two categories to eliminate attacks and protect web servers of the defense sector which encounters different web traffic compared to other sectors (such as e-commerce, web app, etc.). The result shows that by using a machine learning method, a higher accuracy rate, and a lower false alarm detection rate can be achieved.

Keywords: anomaly detection, HTTP protocol, logs, cyber attack, deep learning

Procedia PDF Downloads 213

Authors: Larisa Liamina, Vasily Lavrushin, Salvatore Inguaggiato

Abstract:

This study presents and summarizes the results of researching the isotopic signature of helium in the deep fluid eastern part of the Southern slope of the Greater Caucasus and the Lesser Caucasus (Azerbaijan and Armenia) for the period from 2010 to 2016. The results of isotope ratios of 3He/4He in 59 samples of the gas phase of geothermal fluids and mud volcanoes are presented. New data have been obtained not only on the isotopic ratios of helium, but also neon and argon. The R/Ra ratio was analyzed along the Ankara-Sevan ophiolite structure. The patterns of lateral variations of the 3He/4He ratio of different geological structural elements of the studied region are revealed.

Keywords: isotopes helium, deep fluids, tectonic structures, Caucasus

Procedia PDF Downloads 46

Authors: R. Nandhini, Gaurab Mudbhari

Abstract:

Ranging from the field of health care to self-driving cars, machine learning and deep learning algorithms have revolutionized the field with the proper utilization of images and visual-oriented data. Segmentation, regression, classification, clustering, dimensionality reduction, etc., are some of the Machine Learning tasks that helped Machine Learning and Deep Learning models to become state-of-the-art models for the field where images are key datasets. Among these tasks, classification and clustering are essential but difficult because of the intricate and high-dimensional characteristics of image data. This finding examines and assesses advanced techniques in supervised classification and unsupervised clustering for image datasets, emphasizing the relative efficiency of Convolutional Neural Networks (CNNs), Vision Transformers (ViTs), Deep Embedded Clustering (DEC), and self-supervised learning approaches. Due to the distinctive structural attributes present in images, conventional methods often fail to effectively capture spatial patterns, resulting in the development of models that utilize more advanced architectures and attention mechanisms. In image classification, we investigated both CNNs and ViTs. One of the most promising models, which is very much known for its ability to detect spatial hierarchies, is CNN, and it serves as a core model in our study. On the other hand, ViT is another model that also serves as a core model, reflecting a modern classification method that uses a self-attention mechanism which makes them more robust as this self-attention mechanism allows them to lean global dependencies in images without relying on convolutional layers. This paper evaluates the performance of these two architectures based on accuracy, precision, recall, and F1-score across different image datasets, analyzing their appropriateness for various categories of images. In the domain of clustering, we assess DEC, Variational Autoencoders (VAEs), and conventional clustering techniques like k-means, which are used on embeddings derived from CNN models. DEC, a prominent model in the field of clustering, has gained the attention of many ML engineers because of its ability to combine feature learning and clustering into a single framework and its main goal is to improve clustering quality through better feature representation. VAEs, on the other hand, are pretty well known for using latent embeddings for grouping similar images without requiring for prior label by utilizing the probabilistic clustering method.

Keywords: machine learning, deep learning, image classification, image clustering

Procedia PDF Downloads 17

Authors: Aicha Koulou, Norelislam El Hami, Nabil Hmina

Abstract:

This paper aims at presenting basic concepts and principles in order to develop a methodology to set up sustainable interoperability within collaborative networks. Definitions and clarifications related to the concept of interoperability and sustainability are given. Interoperability levels and cycle that are components supporting the methodology are presented; a structured approach and related phases are proposed.

Keywords: Interoperability, sustainability, collaborative networks, sustainable Interoperability

Procedia PDF Downloads 149

Authors: Erick C. Valverde, Wansu Lim

Abstract:

Facial emotion recognition (FER) system has been recently developed for more advanced computer vision applications. The ability to identify human emotions would enable smart healthcare facility to diagnose mental health illnesses (e.g., depression and stress) as well as better human social interactions with smart technologies. The FER system involves two steps: 1) face detection task and 2) facial emotion recognition task. It classifies the human expression in various categories such as angry, disgust, fear, happy, sad, surprise, and neutral. This system requires intensive research to address issues with human diversity, various unique human expressions, and variety of human facial features due to age differences. These issues generally affect the ability of the FER system to detect human emotions with high accuracy. Early stage of FER systems used simple supervised classification task algorithms like K-nearest neighbors (KNN) and artificial neural networks (ANN). These conventional FER systems have issues with low accuracy due to its inefficiency to extract significant features of several human emotions. To increase the accuracy of FER systems, deep learning (DL)-based methods, like convolutional neural networks (CNN), are proposed. These methods can find more complex features in the human face by means of the deeper connections within its architectures. However, the inference speed and computational costs of a DL-based FER system is often disregarded in exchange for higher accuracy results. To cope with this drawback, an optimized DL-based FER system is proposed in this study.An extreme version of Inception V3, known as Xception model, is leveraged by applying different network optimization methods. Specifically, network pruning and quantization are used to enable lower computational costs and reduce memory usage, respectively. To support low resource requirements, a 68-landmark face detector from Dlib is used in the early step of the FER system.Furthermore, a DL compiler is utilized to incorporate advanced optimization techniques to the Xception model to improve the inference speed of the FER system. In comparison to VGG-Net and ResNet50, the proposed optimized DL-based FER system experimentally demonstrates the objectives of the network optimization methods used. As a result, the proposed approach can be used to create an efficient and real-time FER system.

Keywords: deep learning, face detection, facial emotion recognition, network optimization methods

Procedia PDF Downloads 120

Authors: Imen Elleuch, Wael Ouarda, Gargouri Bilel

Abstract:

We proposes in this paper a Deep Learning based approach to classify text in order to enrich an Arabic ontology based on the objects classes of Gaston Gross. Those object classes are defined by taking into account the syntactic and semantic features of the treated language. Thus, our proposed approach is a hybrid one. In fact, it is based on the one hand on the object classes that represents a knowledge based-approach on classification of text and in the other hand it uses the deep learning approach that use the word embedding-based-approach to classify text. We have applied our proposed approach on a corpus constructed from an Arabic dictionary. The obtained semantic classification of text will enrich the Arabic objects classes ontology. In fact, new classes can be added to the ontology or an expansion of the features that characterizes each object class can be updated. The obtained results are compared to a similar work that treats the same object with a classical linguistic approach for the semantic classification of text. This comparison highlight our hybrid proposed approach that can be ameliorated by broaden the dataset used in the deep learning process.

Keywords: deep-learning approach, object-classes, semantic classification, Arabic

Procedia PDF Downloads 88

Authors: Boris Kolev, Matt Kokan, Mohammad Deriszadeh, Farshid Bateni

Abstract:

Due to the increasing demand for real estate and the need for efficient land utilization in Greater Vancouver, developers have been increasingly considering the construction of high-rise structures with multiple below-grade parking. The temporary excavations required to allow for the construction of underground levels have recently reached up to 40 meters in depth. One of the challenges with deep excavations is the prediction of wall displacements and ground settlements due to their effect on the integrity of City utilities, infrastructure, and adjacent buildings. A large database of survey monitoring data has been collected for deep excavations in various soil conditions and shoring systems. The majority of the data collected is for tie-back anchors and shotcrete lagging systems. The data were categorized, analyzed and the results were evaluated to find a relationship between the most dominant parameters controlling the displacement, such as depth of excavation, soil properties, and the tie-back anchor loading and arrangement. For a select number of deep excavations, finite element modeling was considered for analyses. The lateral displacements from the simulation results were compared to the recorded survey monitoring data. The study concludes with a discussion and comparison of the available empirical and numerical modeling methodologies for evaluating lateral displacements in deep excavations.

Keywords: deep excavations, lateral displacements, numerical modeling, shoring walls, tieback anchors

Procedia PDF Downloads 183

Authors: S. Mousavian, A. Abedianpour, A. Khanmohammadi, S. Hematian, Gh. Eidi Veisi

Abstract:

Ionic liquids are finding a wide range of applications from reaction media to separations and materials processing. In these applications, Vapor–Liquid equilibrium (VLE) is the most important one. VLE for six systems at 353 K and activity coefficients at infinite dilution 〖(γ〗_i^∞) for various solutes (alkanes, alkenes, cycloalkanes, cycloalkenes, aromatics, alcohols, ketones, esters, ethers, and water) in the ionic liquids (1-ethyl-3-methylimidazolium bis (trifluoromethylsulfonyl)imide [EMIM][BTI], 1-hexyl-3-methyl imidazolium bis (trifluoromethylsulfonyl) imide [HMIM][BTI], 1-octyl-3-methylimidazolium bis(trifluoromethylsulfonyl) imide [OMIM][BTI], and 1-butyl-1-methylpyrrolidinium bis (trifluoromethylsulfonyl) imide [BMPYR][BTI]) have been used to train neural networks in the temperature range from (303 to 333) K. Densities of the ionic liquids, Hildebrant constant of substances, and temperature were selected as input of neural networks. The networks with different hidden layers were examined. Networks with seven neurons in one hidden layer have minimum error and good agreement with experimental data.

Keywords: ionic liquid, neural networks, VLE, dilute solution

Procedia PDF Downloads 301

Authors: Mhaned Oubounyt, Jan Baumbach

Abstract:

Differences in co-expression networks between two or multiple cells (sub)types across conditions is a pressing problem in single-cell RNA sequencing (scRNA-seq). A key challenge is to define those co-variations that differ between or among cell types and/or conditions and phenotypes to examine small regulatory networks that can explain mechanistic differences. To this end, we developed SCANet, an all-in-one Python package that uses state-of-the-art algorithms to facilitate the workflow of a combined single-cell GCN (Gene Correlation Network) and GRN (Gene Regulatory Networks) pipeline, including inference of gene co-expression modules from scRNA-seq, followed by trait and cell type associations, hub gene detection, co-regulatory networks, and drug-gene interactions. In an example case, we illustrate how SCANet can be applied to identify regulatory drivers behind a cytokine storm associated with mortality in patients with acute respiratory illness. SCANet is available as a free, open-source, and user-friendly Python package that can be easily integrated into systems biology pipelines.

Keywords: single-cell, co-expression networks, drug-gene interactions, co-regulatory networks

Procedia PDF Downloads 153

Authors: Mohammad R. Jafari, Francois G. Bernardeau

Abstract:

With its arid climate, Qatar experiences low annual rainfall, intense storms, and high evaporation rates. However, the fast-paced rate of infrastructure development in the capital city of Doha has led to recurring instances of surface water flooding as well as rising groundwater levels. Public Work Authority (PWA/ASHGHAL) has implemented an approach to collect and discharge the flood water into a) positive gravity systems; b) Emergency Flooding Area (EFA) – Evaporation, Infiltration or Storage off-site using tankers; and c) Discharge to deep injection wells. As part of the flood prevention scheme, 21 deep injection wells have been constructed to discharge the collected surface and groundwater table in Doha city. These injection wells function as an alternative in localities that do not possess either positive gravity systems or downstream networks that can accommodate additional loads. These injection wells are 400-m deep and are constructed in a complex karstic subsurface condition with large cavities. The injection well system will discharge collected groundwater and storm surface runoff into the permeable Umm Er Radhuma Formation, which is an aquifer present throughout the Persian Gulf Region. The Umm Er Radhuma formation contains saline water that is not being used for water supply. The injection zone is separated by an impervious gypsum formation which acts as a barrier between upper and lower aquifer. State of the art drilling, grouting, and geophysical techniques have been implemented in construction of the wells to assure that the shallow aquifer would not be contaminated and impacted by injected water. Injection and pumping tests were performed to evaluate injection well functionality (injectability). The results of these tests indicated that majority of the wells can accept injection rate of 200 to 300 m³ /h (56 to 83 l/s) under gravity with average value of 250 m³ /h (70 l/s) compared to design value of 50 l/s. This paper presents design and construction process and issues associated with these injection wells, performing injection/pumping tests to determine capacity and effectiveness of the injection wells, the detailed design of collection system and conveying system into the injection wells, and the operation and maintenance process. This system is completed now and is under operation, and therefore, construction of injection wells is an effective option for flood control.

Keywords: deep injection well, flood prevention scheme, geophysical tests, pumping and injection tests, wellhead assembly

Procedia PDF Downloads 120

Authors: Mohammad Yahaghifar

Abstract:

Nowadays Wireless system designers have been facing the continuously increasing demand for high data rates and mobility required by new wireless applications. Evolving future communication network generation cellular wireless networks are envisioned to overcome the fundamental challenges of existing cellular networks, for example, higher data rates, excellent end-to-end performance, and user coverage in hot-spots and crowded areas with lower latency,energy consumption and cost per information transfer. In this paper we propose a potential cellular architecture that separates indoor and outdoor scenarios and discuss various promising technologies for future wireless communication systemssystems, such as massive MIMO, energy-efficient communications,cognitive radio networks, and visible light communications and we disscuse about 5G that is next generation of wireless networks.

Keywords: future challenges in networks, cellur architecture, visible light communication, 5G wireless technologies, spatial modulation, massiva mimo, cognitive radio network, green communications

Procedia PDF Downloads 489

Authors: Mahdi Pirmoradian, Olayinka Adigun, Christos Politis

Abstract:

Cooperative spectrum sensing is a crucial challenge in cognitive radio networks. Cooperative sensing can increase the reliability of spectrum hole detection, optimize sensing time and reduce delay in cooperative networks. In this paper, an efficient central capacity optimization algorithm is proposed to minimize cooperative sensing time in a homogenous sensor network using OR decision rule subject to the detection and false alarm probabilities constraints. The evaluation results reveal significant improvement in the sensing time and normalized capacity of the cognitive sensors.

Keywords: cooperative networks, normalized capacity, sensing time

Procedia PDF Downloads 636

Authors: Selvavinayagan Babiharan

Abstract:

This research aims to develop a deep learning platform to enhance mathematics education for secondary students in Sri Lanka. The platform will be designed to incorporate interactive and user-friendly features to engage students in active learning and promote their mathematical skills. The proposed platform will be developed using TensorFlow and Keras, two widely used deep learning frameworks. The system will be trained on a large dataset of math problems, which will be collected from Sri Lankan school curricula. The results of this research will contribute to the improvement of mathematics education in Sri Lanka and provide a valuable tool for teachers to enhance the learning experience of their students.

Keywords: information technology, education, machine learning, mathematics

Procedia PDF Downloads 84

Authors: Ole-Christian Galbo Engstrøm, Erik Schou Dreier, Birthe Møller Jespersen, Kim Steenstrup Pedersen

Abstract:

Grain quality analysis is a multi-parameterized problem that includes a variety of qualitative and quantitative parameters such as grain type classification, damage type classification, and nutrient regression. Currently, these parameters require human inspection, a multitude of instruments employing a variety of sensor technologies, and predictive model types or destructive and slow chemical analysis. This paper investigates the feasibility of applying near-infrared hyperspectral imaging (NIR-HSI) to grain quality analysis. For this study two datasets of NIR hyperspectral images in the wavelength range of 900 nm - 1700 nm have been used. Both datasets contain images of sparsely and densely packed grain kernels. The first dataset contains ~87,000 image crops of bulk wheat samples from 63 harvests where protein value has been determined by the FOSS Infratec NOVA which is the golden industry standard for protein content estimation in bulk samples of cereal grain. The second dataset consists of ~28,000 image crops of bulk grain kernels from seven different wheat varieties and a single rye variety. In the first dataset, protein regression analysis is the problem to solve while variety classification analysis is the problem to solve in the second dataset. Deep convolutional neural networks (CNNs) have the potential to utilize spatio-spectral correlations within a hyperspectral image to simultaneously estimate the qualitative and quantitative parameters. CNNs can autonomously derive meaningful representations of the input data reducing the need for advanced preprocessing techniques required for classical chemometric model types such as artificial neural networks (ANNs) and partial least-squares regression (PLS-R). A comparison between different CNN architectures utilizing 2D and 3D convolution is conducted. These results are compared to the performance of ANNs and PLS-R. Additionally, a variety of preprocessing techniques from image analysis and chemometrics are tested. These include centering, scaling, standard normal variate (SNV), Savitzky-Golay (SG) filtering, and detrending. The results indicate that the combination of NIR-HSI and CNNs has the potential to be the foundation for an automatic system unifying qualitative and quantitative grain quality analysis within a single sensor technology and predictive model type.

Keywords: deep learning, grain analysis, hyperspectral imaging, preprocessing techniques

Procedia PDF Downloads 100

Authors: Jacob Green, Cecilia Cabrera, Maximilian Jakobs, Andrea Dimitracopoulos, Mark van der Wilk, Ryan Greenhalgh

Abstract:

Predicting drug properties is key in drug discovery to enable de-risking of assets before expensive clinical trials and to find highly active compounds faster. Interest from the machine learning community has led to the release of a variety of benchmark datasets and proposed methods. However, it remains unclear for practitioners which method or approach is most suitable, as different papers benchmark on different datasets and methods, leading to varying conclusions that are not easily compared. Our large-scale empirical study links together numerous earlier works on different datasets and methods, thus offering a comprehensive overview of the existing property classes, datasets, and their interactions with different methods. We emphasise the importance of uncertainty quantification and the time and, therefore, cost of applying these methods in the drug development decision-making cycle. To the best of the author's knowledge, it has been observed that the optimal approach varies depending on the dataset and that engineered features with classical machine learning methods often outperform deep learning. Specifically, QSAR datasets are typically best analysed with classical methods such as Gaussian Processes, while ADMET datasets are sometimes better described by Trees or deep learning methods such as Graph Neural Networks or language models. Our work highlights that practitioners do not yet have a straightforward, black-box procedure to rely on and sets a precedent for creating practitioner-relevant benchmarks. Deep learning approaches must be proven on these benchmarks to become the practical method of choice in drug property prediction.

Keywords: activity (QSAR), ADMET, classical methods, drug property prediction, empirical study, machine learning

Procedia PDF Downloads 83

Authors: Degale Desta, Cheng Jian

Abstract:

The most important factor in identifying a person is their face. Even identical twins have their own distinct faces. As a result, identification and face recognition are needed to tell one person from another. A face recognition system is a verification tool used to establish a person's identity using biometrics. Nowadays, face recognition is a common technique used in a variety of applications, including home security systems, criminal identification, and phone unlock systems. This system is more secure because it only requires a facial image instead of other dependencies like a key or card. Face detection and face identification are the two phases that typically make up a human recognition system.The idea behind designing and creating a face recognition system using deep learning with Azure ML Python's OpenCV is explained in this paper. Face recognition is a task that can be accomplished using deep learning, and given the accuracy of this method, it appears to be a suitable approach. To show how accurate the suggested face recognition system is, experimental results are given in 98.46% accuracy using Fast-RCNN Performance of algorithms under different training conditions.

Keywords: deep learning, face recognition, identification, fast-RCNN

Procedia PDF Downloads 140

Authors: Donghee Noh, Seonhyeong Kim, Junhwan Choi, Heegon Kim, Sooho Jung, Keunho Park

Abstract:

In this paper, we describe an image post-processing method to enhance the performance of the parcel segmentation method using deep learning-based aerial images conducted in previous studies. The study results were evaluated using a confusion matrix, IoU, Precision, Recall, and F1-Score. In the case of the confusion matrix, it was observed that the false positive value, which is the result of misclassification, was greatly reduced as a result of image post-processing. The average IoU was 0.9688 in the image post-processing, which is higher than the deep learning result of 0.8362, and the F1-Score was also 0.9822 in the image post-processing, which was higher than the deep learning result of 0.8850. As a result of the experiment, it was found that the proposed technique positively complements the deep learning results in segmenting the parcel of interest.

Keywords: aerial image, image process, machine vision, open field smart farm, segmentation

Procedia PDF Downloads 82

Authors: Aisulu Tileukulova, Zhexebay Dauren

Abstract:

Radar plays an important role because it is widely used in civil and military fields. Target detection is one of the most important radar applications. The accuracy of detecting inconspicuous aerial objects in radar facilities is lower against the background of noise. Convolutional neural networks can be used to improve the recognition of this type of aerial object. The purpose of this work is to develop an algorithm for recognizing aerial objects using convolutional neural networks, as well as training a neural network. In this paper, the structure of a convolutional neural network (CNN) consists of different types of layers: 8 convolutional layers and 3 layers of a fully connected perceptron. ReLU is used as an activation function in convolutional layers, while the last layer uses softmax. It is necessary to form a data set for training a neural network in order to detect a target. We built a Confusion Matrix of the CNN model to measure the effectiveness of our model. The results showed that the accuracy when testing the model was 95.7%. Classification of echo signals using CNN shows high accuracy and significantly speeds up the process of predicting the target.

Keywords: radar, neural network, convolutional neural network, echo signals

Procedia PDF Downloads 354

Authors: Sorin Valcan, Mihail Gaianu

Abstract:

Labeling is a very costly and time consuming process which aims to generate datasets for training neural networks in several functionalities and projects. For driver monitoring system projects, the need for labeled images has a significant impact on the budget and distribution of effort. This paper presents the modifications done to an algorithm used for the generation of ground truth data for 2D eyes location on infrared images with drivers in order to improve the quality of the data and performance of the trained neural networks. The algorithm restrictions become tougher, which makes it more accurate but also less constant. The resulting dataset becomes smaller and shall not be altered by any kind of manual label adjustment before being used in the neural networks training process. These changes resulted in a much better performance of the trained neural networks.

Keywords: labeling automation, infrared camera, driver monitoring, eye detection, convolutional neural networks

Procedia PDF Downloads 119

Authors: Edgardo V. Gubatanga Jr., Mark Joshua Salvacion

Abstract:

Most aerial LiDAR systems have accompanying aerial cameras in order to capture not only the terrain of the surveyed area but also its true-color appearance. However, the presence of atmospheric clouds, poor lighting conditions, and aerial camera problems during an aerial survey may cause absence of aerial photographs. These leave areas having terrain information but lacking aerial photographs. Intensity images can be derived from LiDAR data but they are only grayscale images. A deep learning model is developed to create a complex function in a form of a deep neural network relating the pixel values of LiDAR-derived intensity images and true-color images. This complex function can then be used to predict the true-color images of a certain area using intensity images from LiDAR data. The predicted true-color images do not necessarily need to be accurate compared to the real world. They are only intended to look realistic so that they can be used as base maps.

Keywords: aerial LiDAR, colorization, deep learning, intensity images

Procedia PDF Downloads 168

Authors: Kumar Divyendra

Abstract:

Underwater Sensor Networks (UWSNs) have wide applications in areas such as water quality monitoring, marine wildlife management etc. A typical UWSN system consists of a set of sensors deployed randomly underwater which communicate with each other using acoustic links. RF communication doesn't work underwater, and GPS too isn't available underwater. Additionally Automated Underwater Vehicles (AUVs) are deployed to collect data from some special nodes called Cluster Heads (CHs). These CHs aggregate data from their neighboring nodes and forward them to the AUVs using optical links when an AUV is in range. This helps reduce the number of hops covered by data packets and helps conserve energy. We consider the three-dimensional model of the UWSN. Nodes are initially deployed randomly underwater. They attach themselves to the surface using a rod and can only move upwards or downwards using a pump and bladder mechanism. We use graph theory concepts to maximize the coverage volume while every node maintaining connectivity with at least one surface node. We treat the surface nodes as landmarks and each node finds out its hop distance from every surface node. We treat these hop-distances as coordinates and use them for AUV navigation. An AUV intending to move closer to a node with given coordinates moves hop by hop through nodes that are closest to it in terms of these coordinates. In absence of GPS, multiple different approaches like Inertial Navigation System (INS), Doppler Velocity Log (DVL), computer vision-based navigation, etc., have been proposed. These systems have their own drawbacks. INS accumulates error with time, vision techniques require prior information about the environment. We propose a method that makes use of the earth's magnetic field values for navigation and combines it with other methods that simultaneously increase the coverage volume under the UWSN. The AUVs are fitted with magnetometers that measure the magnetic intensity (I), horizontal inclination (H), and Declination (D). The International Geomagnetic Reference Field (IGRF) is a mathematical model of the earth's magnetic field, which provides the field values for the geographical coordinateson earth. Researchers have developed an inverse deep learning model that takes the magnetic field values and predicts the location coordinates. We make use of this model within our work. We combine this with with the hop-by-hop movement described earlier so that the AUVs move in such a sequence that the deep learning predictor gets trained as quickly and precisely as possible We run simulations in MATLAB to prove the effectiveness of our model with respect to other methods described in the literature.

Keywords: clustering, deep learning, network backbone, parallel computing

Procedia PDF Downloads 99

Authors: Nabil Shawkat, Jamil Saquer

Abstract:

Social media has accelerated our ability to engage with others and eliminated many communication barriers. On the other hand, the widespread use of social media resulted in an increase in online hate speech. This has drastic impacts on vulnerable individuals and societies. Therefore, it is critical to detect hate speech to prevent innocent users and vulnerable communities from becoming victims of hate speech. We investigate the performance of different deep learning and machine learning algorithms on three different datasets. Our results show that the BERT model gives the best performance among all the models by achieving an F1-score of 90.6% on one of the datasets and F1-scores of 89.7% and 88.2% on the other two datasets.

Keywords: hate speech, machine learning, deep learning, abusive words, social media, text classification

Procedia PDF Downloads 139