Search results for: computer assisted classification

Authors: Sirilak Areerachakul

Abstract:

Water quality has initiated serious management efforts in many countries. Artificial Neural Network (ANN) models are developed as forecasting tools in predicting water quality trend based on historical data. This study endeavors to automatically classify water quality. The water quality classes are evaluated using 6 factor indices. These factors are pH value (pH), Dissolved Oxygen (DO), Biochemical Oxygen Demand (BOD), Nitrate Nitrogen (NO3N), Ammonia Nitrogen (NH3N) and Total Coliform (T-Coliform). The methodology involves applying data mining techniques using multilayer perceptron (MLP) neural network models. The data consisted of 11 sites of Saen Saep canal in Bangkok, Thailand. The data is obtained from the Department of Drainage and Sewerage Bangkok Metropolitan Administration during 2007-2011. The results of multilayer perceptron neural network exhibit a high accuracy multilayer perception rate at 94.23% in classifying the water quality of Saen Saep canal in Bangkok. Subsequently, this encouraging result could be combined with GIS data improves the classification accuracy significantly.

Keywords: artificial neural network, geographic information system, water quality, computer science

Procedia PDF Downloads 343

Authors: Bruno R. A. Oliveira, Italo V. V. Braga, Jonas P. Reges, Luiz P. O. Santos, Sidney C. Duarte, Emilson R. R. Melo, Auzuir R. Alexandria

Abstract:

This project aims to building a controlled greenhouse, or for better understanding, a structure where one can maintain a given range of temperature values (°C) coming from radiation emitted by an incandescent light, as previously defined, characterizing as a kind of on-off control and a differential, which is the plotting of temperature versus time graphs assisted by MATLAB software via serial communication. That way it is possible to connect the stove with a computer and monitor parameters. In the control, it was performed using a PIC 16F877A microprocessor which enabled convert analog signals to digital, perform serial communication with the IC MAX232 and enable signal transistors. The language used in the PIC's management is Basic. There are also a cooling system realized by two coolers 12V distributed in lateral structure, being used for venting and the other for exhaust air. To find out existing temperature inside is used LM35DZ sensor. Other mechanism used in the greenhouse construction was comprised of a reed switch and a magnet; their function is in recognition of the door position where a signal is sent to a buzzer when the door is open. Beyond it exist LEDs that help to identify the operation which the stove is located. To facilitate human-machine communication is employed an LCD display that tells real-time temperature and other information. The average range of design operating without any major problems, taking into account the limitations of the construction material and structure of electrical current conduction, is approximately 65 to 70 ° C. The project is efficient in these conditions, that is, when you wish to get information from a given material to be tested at temperatures not as high. With the implementation of the greenhouse automation, facilitating the temperature control and the development of a structure that encourages correct environment for the most diverse applications.

Keywords: greenhouse, microcontroller, temperature, control, MATLAB

Procedia PDF Downloads 402

Authors: Hossein Alimohammadi, Mohsen Amirmojahedi, Mehrdad Rowhani

Abstract:

Soil properties are used by geotechnical engineers to evaluate and analyze site conditions for designing purposes. Although basic soil classification tests are easy to perform and provide useful information to determine the properties of soils, it may take time to get the result and add some costs to the projects. Standard Penetration Test (SPT) provides an opportunity to evaluate soil parameters without performing laboratory tests. In addition to its simplicity and cheapness, the results become available immediately. This research provides a guideline on the application of the SPT test method, reliability of adapting the SPT test results in evaluating soil physical and mechanical properties such as Atterberg limits, shear strength, and compressive strength compressibility parameters. A total of 70 boreholes were investigated in this study by taking soil samples between depths of 1.2 to 15.25 meters. The project site was located in Morrow County, Ohio. A regression-based formula was proposed based on Tobit regression with a stepwise variable selection analysis conducted between SPT and other typical soil properties obtained from soil tests. The results of the research illustrated that the shear strength and physical properties of the soil affect the SPT number. The proposed correlation can help engineers to use SPT test results in their design with higher accuracy.

Keywords: standard penetration test, soil properties, soil classification, regression method

Procedia PDF Downloads 188

Authors: Abdul Ghaffar Memon, Xiao Hong Zhou, Yunpeng Xing, Ruoyu Wang, Miao He

Abstract:

Due to deleterious environmental and health effects of the Hg²⁺ ions, various online, detection methods apart from the traditional analytical tools have been developed by researchers. Biosensors especially, label, label-free, colorimetric and optical sensors have advanced with sensitive detection. However, there remains a gap of ultrasensitive quantification as noise interact significantly especially in the AuNP based label-free colorimetry. This study reported an amplification strategy using Exo-III enzyme for target recycling of Hg²⁺ ions in a T-rich hairpin loop metallobase label-free colorimetric nanosensor with an improved sensitivity using unmodified gold nanoparticles (uGNPs) as an indicator. The two T-rich metallobase hairpin loop structures as 5’- CTT TCA TAC ATA GAA AAT GTA TGT TTG -3 (HgS1), and 5’- GGC TTT GAG CGC TAA GAA A TA GCG CTC TTT G -3’ (HgS2) were tested in the study. The thermodynamic properties of HgS1 and HgS2 were calculated using online tools (http://biophysics.idtdna.com/cgi-bin/meltCalculator.cgi). The lab scale synthesized uGNPs were utilized in the analysis. The DNA sequence had T-rich bases on both tails end, which in the presence of Hg²⁺ forms a T-Hg²⁺-T mismatch, promoting the formation of dsDNA. Later, the Exo-III incubation enable the enzyme to cleave stepwise mononucleotides from the 3’ end until the structure become single-stranded. These ssDNA fragments then adsorb on the surface of AuNPs in their presence and protect AuNPs from the induced salt aggregation. The visible change in color from blue (aggregation stage in the absence of Hg²⁺) and pink (dispersion state in the presence of Hg²⁺ and adsorption of ssDNA fragments) can be observed and analyzed through UV spectrometry. An ultrasensitive quantitative nanosensor employing Exo-III assisted target recycling of mercury ions through label-free colorimetry with nanomolar detection using uGNPs have been achieved and is further under the optimization to achieve picomolar range by avoiding the influence of the environmental matrix. The proposed strategy will supplement in the direction of uGNP based ultrasensitive, rapid, onsite, label-free colorimetric detection.

Keywords: colorimetric, Exo-III, gold nanoparticles, Hg²⁺ detection, label-free, signal amplification

Procedia PDF Downloads 311

Authors: Sukhleen Kaur

Abstract:

In security groundwork, Intrusion Detection System (IDS) has become an important component. The IDS has received increasing attention in recent years. IDS is one of the effective way to detect different kinds of attacks and malicious codes in a network and help us to secure the network. Data mining techniques can be implemented to IDS, which analyses the large amount of data and gives better results. Data mining can contribute to improving intrusion detection by adding a level of focus to anomaly detection. So far the study has been carried out on finding the attacks but this paper detects the malicious files. Some intruders do not attack directly, but they hide some harmful code inside the files or may corrupt those file and attack the system. These files are detected according to some defined parameters which will form two lists of files as normal files and harmful files. After that data mining will be performed. In this paper a hybrid classifier has been used via Naive Bayes and Ripper classification methods. The results show how the uploaded file in the database will be tested against the parameters and then it is characterised as either normal or harmful file and after that the mining is performed. Moreover, when a user tries to mine on harmful file it will generate an exception that mining cannot be made on corrupted or harmful files.

Keywords: data mining, association, classification, clustering, decision tree, intrusion detection system, misuse detection, anomaly detection, naive Bayes, ripper

Procedia PDF Downloads 414

Authors: Marcos Silva Tavares, Jamile Raquel Regazzo, Edson José de Souza Sardinha, Murilo Mesquita Baesso

Abstract:

Common beans are a widely cultivated and consumed legume globally, serving as a staple food for humans, especially in developing countries, due to their nutritional characteristics. Nitrogen (N) is the most limiting nutrient for productivity, and foliar analysis is crucial to ensure balanced nitrogen fertilization. Excessive N applications can cause, either isolated or cumulatively, soil and water contamination, plant toxicity, and increase their susceptibility to diseases and pests. However, the quantification of N using conventional methods is time-consuming and costly, demanding new technologies to optimize the adequate supply of N to plants. Thus, it becomes necessary to establish constant monitoring of the foliar content of this macronutrient in plants, mainly at the V4 stage, aiming at precision management of nitrogen fertilization. In this work, the objective was to evaluate the performance of a deep learning model, Resnet-50, in the classification of foliar nitrogen in common beans using RGB images. The BRS Estilo cultivar was sown in a greenhouse in a completely randomized design with four nitrogen doses (T1 = 0 kg N ha-1, T2 = 25 kg N ha-1, T3 = 75 kg N ha-1, and T4 = 100 kg N ha-1) and 12 replications. Pots with 5L capacity were used with a substrate composed of 43% soil (Neossolo Quartzarênico), 28.5% crushed sugarcane bagasse, and 28.5% cured bovine manure. The water supply of the plants was done with 5mm of water per day. The application of urea (45% N) and the acquisition of images occurred 14 and 32 days after sowing, respectively. A code developed in Matlab© R2022b was used to cut the original images into smaller blocks, originating an image bank composed of 4 folders representing the four classes and labeled as T1, T2, T3, and T4, each containing 500 images of 224x224 pixels obtained from plants cultivated under different N doses. The Matlab© R2022b software was used for the implementation and performance analysis of the model. The evaluation of the efficiency was done by a set of metrics, including accuracy (AC), F1-score (F1), specificity (SP), area under the curve (AUC), and precision (P). The ResNet-50 showed high performance in the classification of foliar N levels in common beans, with AC values of 85.6%. The F1 for classes T1, T2, T3, and T4 was 76, 72, 74, and 77%, respectively. This study revealed that the use of RGB images combined with deep learning can be a promising alternative to slow laboratory analyses, capable of optimizing the estimation of foliar N. This can allow rapid intervention by the producer to achieve higher productivity and less fertilizer waste. Future approaches are encouraged to develop mobile devices capable of handling images using deep learning for the classification of the nutritional status of plants in situ.

Keywords: convolutional neural network, residual network 50, nutritional status, artificial intelligence

Procedia PDF Downloads 19

Authors: Brendah Nagula, Omolo Fredrick Okalebo, Ronald Ssengendo, Ivan Bamweyana

Abstract:

The Kampala Physical Development Plan (KPDP) was developed in 2012 and projected both long term and short term developments within the City .The purpose of the plan was to not only shape the city into a spatially planned area but also to control the urban sprawl trends that had expanded with pronounced instances of informal settlements. This plan was approved by the National Physical Planning Board and a signature was appended by the Minister in 2013. Much as the KPDP plan has been implemented using different approaches such as detailed planning, development control, subdivision planning, carrying out construction inspections, greening and beautification, there is still limited knowledge on the level of conformance towards this plan. Therefore, it is yet to be determined whether it has been effective in shaping the City into an ideal spatially planned area. Attaining a clear picture of the level of conformance towards the KPDP 2012 through evaluation between the planned and the existing land use in Kampala City was performed. Methods such as Supervised Classification and Post Classification Change Detection were adopted to perform this evaluation. Scrutiny of findings revealed Central Division registered the lowest level of conformance to the planning standards specified in the KPDP 2012 followed by Nakawa, Rubaga, Kawempe, and Makindye. Furthermore, mixed-use development was identified as the land use with the highest level of non-conformity of 25.11% and institutional land use registered the highest level of conformance of 84.45 %. The results show that the aspect of location was not carefully considered while allocating uses in the KPDP whereby areas located near the Central Business District have higher land rents and hence require uses that ensure profit maximization. Also, the prominence of development towards mixed-use denotes an increased demand for land towards compact development that was not catered for in the plan. Therefore in order to transform Kampala city into a spatially planned area, there is need to carefully develop detailed plans especially for all the Central Division planning precincts indicating considerations for land use densification.

Keywords: spatial plan, post classification change detection, Kampala city, landuse

Procedia PDF Downloads 92

Authors: Babak Sadeghi, Rony Snyders, Marie-Paule.Delplancke-Ogletree

Abstract:

Nitrogen fixation (NF) is one of the crucial industrial processes. Many attempts have been made in order to artificially fix nitrogen, and among them, the Haber-Bosch’s (H-B) process is widely used. However, it presents two major drawbacks: huge fossil feedstock consumption and noticeable greenhouse gases emission. It is, therefore, necessary to develop alternatives. Plasma technology, as an inherent “green” technology, is considered to have a great potential for reducing the environmental impacts and improving the energy efficiency of the NF process. In this work, we have studied the catalyst assisted microwave plasma for NF application. Heterogeneous catalysts of MoO₃, with various loads 0, 5, 10, 20, and 30 wt%, supported on γ-alumina were prepared by conventional wet impregnation. Crystallinity, surface area, pore size, and microstructure were obtained by X-ray diffraction (XRD), Brunauer–Emmett–Teller (BET) adsorption isotherm, Scanning electron microscopy (SEM), and Transmission electron microscopy (TEM). The XRD patterns of calcined alumina confirm the γ- phase. Characteristic picks of MoO₃ could not be observed for low loads (< 20 wt%), likely indicating a high dispersion of metal oxide over the support. The specific surface area along with pores size are decreasing with increasing calcination temperature and MoO₃ loading. The MoO₃ loading does not modify the microstructure. TEM and SEM results for loading inferior to 20 wt% are coherent with a monolayer of MoO₃ on the support as proposed elsewhere. For loading of 20 wt% and more, TEM and Electron diffraction (ED) show nanocrystalline ₃-D MoO₃ particles. The catalytic performances of these catalysts were investigated in the post-discharge of a microwave plasma for NOx formation from N₂/O₂ mixtures. The plasma is sustained by a surface wave launched in a quartz tube via a surfaguide supplied by a 2.45 GHz microwave generator in pulse mode. In-situ identification and quantification of the products were carried out by Fourier-transform infrared spectroscopy (FTIR) in the post-discharge region. FTIR analysis of the exhausted gas reveal NO and NO₂ bands in presence of catalyst while only NO band were assigned without catalyst. On the other hand, in presence of catalyst, a 10% increase of NOₓ formation and of 20% increase in energy efficiency are observed.

Keywords: γ-Al2O₃-MoO₃, µ-waveplasma, N2 fixation, Plasma-catalysis, Plasma diagnostic

Procedia PDF Downloads 176

Authors: Aspen W. Glaspell, Victoria J. Rouse, Brian K. Friedrich, Kyosung Choo

Abstract:

Heat transfer of two-phase impinging jet on a flat plate surface are experimentally investigated. The effects of the nozzle-to-surface distance and volumetric quality on the Nusselt number are considered. The results show that the normalized stagnation Nusselt number drastically increase with decreasing the nozzle-to-surface distance due to the jet deflection effect. Based on the experimental results, new correlations for the stagnation Nusselt number are developed as a function of the nozzle-to-surface distance.

Keywords: jet impingement, water jet, air assisted, circular jet

Procedia PDF Downloads 191

Authors: Yavuz Unal, Kemal Polat, H. Erdinc Kocer

Abstract:

In this study, a decision support system comprising three stages has been proposed to detect the disc abnormalities of the lumbar region. In the first stage named the feature extraction, T2-weighted sagittal and axial Magnetic Resonance Images (MRI) were taken from 55 people and then 27 appearance and shape features were acquired from both sagittal and transverse images. In the second stage named the feature weighting process, k-means clustering based feature weighting (KMCBFW) proposed by Gunes et al. Finally, in the third stage named the classification process, the classifier algorithms including multi-layer perceptron (MLP- neural network), support vector machine (SVM), Naïve Bayes, and decision tree have been used to classify whether the subject has lumbar disc or not. In order to test the performance of the proposed method, the classification accuracy (%), sensitivity, specificity, precision, recall, f-measure, kappa value, and computation times have been used. The best hybrid model is the combination of k-means clustering based feature weighting and decision tree in the detecting of lumbar disc disease based on both sagittal and axial MR images.

Keywords: lumbar disc abnormality, lumbar MRI, lumbar spine, hybrid models, hybrid features, k-means clustering based feature weighting

Procedia PDF Downloads 520

Authors: Talal Alshammari, Nasser Alshammari, Mohamed Sedky, Chris Howard

Abstract:

With the widespread adoption of the Internet-connected devices, and with the prevalence of the Internet of Things (IoT) applications, there is an increased interest in machine learning techniques that can provide useful and interesting services in the smart home domain. The areas that machine learning techniques can help advance are varied and ever-evolving. Classifying smart home inhabitants’ Activities of Daily Living (ADLs), is one prominent example. The ability of machine learning technique to find meaningful spatio-temporal relations of high-dimensional data is an important requirement as well. This paper presents a comparative evaluation of state-of-the-art machine learning techniques to classify ADLs in the smart home domain. Forty-two synthetic datasets and two real-world datasets with multiple inhabitants are used to evaluate and compare the performance of the identified machine learning techniques. Our results show significant performance differences between the evaluated techniques. Such as AdaBoost, Cortical Learning Algorithm (CLA), Decision Trees, Hidden Markov Model (HMM), Multi-layer Perceptron (MLP), Structured Perceptron and Support Vector Machines (SVM). Overall, neural network based techniques have shown superiority over the other tested techniques.

Keywords: activities of daily living, classification, internet of things, machine learning, prediction, smart home

Procedia PDF Downloads 357

Authors: Sarai Guerrero, Lijia Liu

Abstract:

Zinc gallogermanate (ZGGO) is a persistent phosphor that can emit in the near infrared (NIR) range once dopped with Cr³⁺ enabling its use for in-vivo deep-tissue bio-imaging. Such a property also allows for its application in cancer diagnosis and therapy. Given this, work into developing a synthetic procedure that can be done using common laboratory instruments and equipment as well as understanding ZGGO overall, is in demand. However, the ZGGO nanoparticles must have a size compatible for cell intake to occur while still maintaining sufficient photoluminescence. The nanoparticle must also be made biocompatible by functionalizing the surface for hydrophilic solubility and for high particle uniformity in the final product. Additionally, most research is completed on doped ZGGO, leaving a gap in understanding the base form of ZGGO. It also leaves a gap in understanding how doping affects the synthesis of ZGGO. In this work, the first step of optimizing the particle size via the crystalline size of ZGGO was done with undoped ZGGO using the organic acid, oleic acid (OA) for organic ligand-assisted biphasic hydrothermal synthesis. The effects of this synthesis procedure on ZGGO’s crystallinity were evaluated using Powder X-Ray Diffraction (PXRD). OA was selected as the capping ligand as experiments have shown it beneficial in synthesizing sub-10 nm zinc gallate (ZGO) nanoparticles as well as palladium nanocrystals and magnetite (Fe₃O₄) nanoparticles. Later it is possible to substitute OA with a different ligand allowing for hydrophilic solubility. Attenuated Total Reflection Fourier-Transform Infrared (ATR-FTIR) was used to investigate the surface of the nanoparticle to investigate and verify that OA had capped the nanoparticle. PXRD results showed that using this procedure led to improved crystallinity, comparable to the high-purity reagents used on the ZGGO nanoparticles. There was also a change in the crystalline size of the ZGGO nanoparticles. ATR-FTIR showed that once capped ZGGO cannot be annealed as doing so will affect the OA. These results point to this new procedure positively affecting the crystallinity of ZGGO nanoparticles. There are also repeatable implying the procedure is a reliable source of highly crystalline ZGGO nanoparticles. With this completed, the next step will be working on substituting the OA with a hydrophilic ligand. As these ligands effect the solubility of the nanoparticle as well as the pH that the nanoparticles can dissolve in, further research is needed to verify which ligand is best suited for preparing ZGGO for bio-imaging.

Keywords: biphasic hydrothermal synthesis, crystallinity, oleic acid, zinc gallogermanate

Procedia PDF Downloads 133

Authors: Jeron Maheswaran, Sudath C. Siriwardane

Abstract:

In fatigue analysis, the structural detail of tubular joint has taken great attention among engineers. The DNV-RP-C203 is covering this topic quite well for simple and clear joint cases. For complex joint and geometry, where joint classification isn’t available and limitation on validity range of non-dimensional geometric parameters, the challenges become a fact among engineers. The classification of joint is important to carry out through the fatigue analysis. These joint configurations are identified by the connectivity and the load distribution of tubular joints. To overcome these problems to some extent, this paper compare the fatigue life of tubular joints in offshore jacket according to the stress concentration factors (SCF) in DNV-RP-C203 and finite element method employed Abaqus/CAE. The paper presents the geometric details, material properties and considered load history of the jacket structure. Describe the global structural analysis and identification of critical tubular joints for fatigue life estimation. Hence fatigue life is determined based on the guidelines provided by design codes. Fatigue analysis of tubular joints is conducted using finite element employed Abaqus/CAE [4] as next major step. Finally, obtained SCFs and fatigue lives are compared and their significances are discussed.

Keywords: fatigue life, stress-concentration factor, finite element analysis, offshore jacket structure

Procedia PDF Downloads 453

Authors: Ana Teixeira, Joao Orvalho

Abstract:

This paper describes experiments using Scratch games, to check the feasibility of employing cerebral palsy users gestures as an alternative of interaction with a computer carried out by students of Master Human Computer Interaction (HCI) of IPC Coimbra. The main focus of this work is to study the usability of a Web Camera as a motion tracking device to achieve a virtual human-computer interaction used by individuals with CP. An approach for Human-computer Interaction (HCI) is present, where individuals with cerebral palsy react and interact with a scratch game through the use of a webcam as an external interaction device. Motion tracking interaction is an emerging technology that is becoming more useful, effective and affordable. However, it raises new questions from the HCI viewpoint, for example, which environments are most suitable for interaction by users with disabilities. In our case, we put emphasis on the accessibility and usability aspects of such interaction devices to meet the special needs of people with disabilities, and specifically people with CP. Despite the fact that our work has just started, preliminary results show that, in general, computer vision interaction systems are very useful; in some cases, these systems are the only way by which some people can interact with a computer. The purpose of the experiments was to verify two hypothesis: 1) people with cerebral palsy can interact with a computer using their natural gestures, 2) scratch games can be a research tool in experiments with disabled young people. A game in Scratch with three levels is created to be played through the use of a webcam. This device permits the detection of certain key points of the user’s body, which allows to assume the head, arms and specially the hands as the most important aspects of recognition. Tests with 5 individuals of different age and gender were made throughout 3 days through periods of 30 minutes with each participant. For a more extensive and reliable statistical analysis, the number of both participants and repetitions in further investigations should be increased. However, already at this stage of research, it is possible to draw some conclusions. First, and the most important, is that simple scratch games on the computer can be a research tool that allows investigating the interaction with computer performed by young persons with CP using intentional gestures. Measurements performed with the assistance of games are attractive for young disabled users. The second important conclusion is that they are able to play scratch games using their gestures. Therefore, the proposed interaction method is promising for them as a human-computer interface. In the future, we plan to include the development of multimodal interfaces that combine various computer vision devices with other input devices improvements in the existing systems to accommodate more the special needs of individuals, in addition, to perform experiments on a larger number of participants.

Keywords: motion tracking, cerebral palsy, rehabilitation, HCI

Procedia PDF Downloads 235

Authors: Omid Kalatpour, Mohammadreza Ajdari

Abstract:

This article studied the possible causes of human error-induced incidents in the petrochemical industry maintenance activities by using Human Factors Analysis and Classification System (HFACS). The purpose of the study was anticipating and identifying these causes and proposing corrective and preventive actions. Maintenance department in a petrochemical company was selected for research. A checklist of human error-induced incidents was developed based on four HFACS main levels and nineteen sub-groups. Hierarchical task analysis (HTA) technique was used to identify maintenance activities and tasks. The main causes of possible incidents were identified by checklist and recorded. Corrective and preventive actions were defined depending on priority. Analyzing the worksheets of 444 activities in four levels of HFACS showed 37.6% of the causes were at the level of unsafe actions, 27.5% at the level of unsafe supervision, 20.9% at the level of preconditions for unsafe acts and 14% of the causes were at the level of organizational effects. The HFACS sub-groups showed errors (24.36%) inadequate supervision (14.89%) and violations (13.26%) with the most frequency. According to findings of this study, increasing the training effectiveness of operators and supervision improvement respectively are the most important measures in decreasing the human error-induced incidents in petrochemical industry maintenance.

Keywords: human error, petrochemical industry, maintenance, HFACS

Procedia PDF Downloads 242

Authors: Fei Gao, Xinyang Dong, Yisu Ge, Shufang Lu, Libo Weng

Abstract:

Vehicle color recognition plays an important role in traffic accident investigation. However, due to the influence of illumination, weather, and noise, vehicle color recognition still faces challenges. In this paper, a vehicle color classification network based on spatial cluster loss and channel attention mechanism (SCNet) is proposed for vehicle color recognition. A channel attention module is applied to extract the features of vehicle color representative regions and reduce the weight of nonrepresentative color regions in the channel. The proposed loss function, called spatial clustering loss (SC-loss), consists of two channel-specific components, such as a concentration component and a diversity component. The concentration component forces all feature channels belonging to the same class to be concentrated through the channel cluster. The diversity components impose additional constraints on the channels through the mean distance coefficient, making them mutually exclusive in spatial dimensions. In the comparison experiments, the proposed method can achieve state-of-the-art performance on the public datasets, VCD, and VeRi, which are 96.1% and 96.2%, respectively. In addition, the ablation experiment further proves that SC-loss can effectively improve the accuracy of vehicle color recognition.

Keywords: feature extraction, convolutional neural networks, intelligent transportation, vehicle color recognition

Procedia PDF Downloads 183

Authors: Nguyen J. M., Lucas G., Ruan S., Digonnet H., Antonioli D.

Abstract:

Introduction: Deep Learning (DL) is a very powerful tool for analyzing image data. But for tabular data, it cannot compete with machine learning methods like XGBoost. The research question becomes: can tabular data be transformed into images that can be analyzed by simple CNNs (Convolutional Neuron Networks)? Will DL be the absolute tool for data classification? All current solutions consist in repositioning the variables in a 2x2 matrix using their correlation proximity. In doing so, it obtains an image whose pixels are the variables. We implement a technology, DeepNIC, that offers the possibility of obtaining an image for each variable, which can be analyzed by simple CNNs. Material and method: The 'ROP' (Regression OPtimized) model is a binary and atypical decision tree whose nodes are managed by a new artificial neuron, the Neurop. By positioning an artificial neuron in each node of the decision trees, it is possible to make an adjustment on a theoretically infinite number of variables at each node. From this new decision tree whose nodes are artificial neurons, we created the concept of a 'Random Forest of Perfect Trees' (RFPT), which disobeys Breiman's concepts by assembling very large numbers of small trees with no classification errors. From the results of the RFPT, we developed a family of 10 statistical information criteria, Nguyen Information Criterion (NICs), which evaluates in 3 dimensions the predictive quality of a variable: Performance, Complexity and Multiplicity of solution. A NIC is a probability that can be transformed into a grey level. The value of a NIC depends essentially on 2 super parameters used in Neurops. By varying these 2 super parameters, we obtain a 2x2 matrix of probabilities for each NIC. We can combine these 10 NICs with the functions AND, OR, and XOR. The total number of combinations is greater than 100,000. In total, we obtain for each variable an image of at least 1166x1167 pixels. The intensity of the pixels is proportional to the probability of the associated NIC. The color depends on the associated NIC. This image actually contains considerable information about the ability of the variable to make the prediction of Y, depending on the presence or absence of other variables. A basic CNNs model was trained for supervised classification. Results: The first results are impressive. Using the GSE22513 public data (Omic data set of markers of Taxane Sensitivity in Breast Cancer), DEEPNic outperformed other statistical methods, including XGBoost. We still need to generalize the comparison on several databases. Conclusion: The ability to transform any tabular variable into an image offers the possibility of merging image and tabular information in the same format. This opens up great perspectives in the analysis of metadata.

Keywords: tabular data, CNNs, NICs, DeepNICs, random forest of perfect trees, classification

Procedia PDF Downloads 125

Authors: Bowei Yuan, Shi Li, Liuyang Song, Huaqing Wang, Lingli Cui

Abstract:

Convolutional neural networks (CNN) have received extensive attention in the field of fault diagnosis. Many fault diagnosis methods use CNN for fault type identification. However, when the amount of raw data collected by sensors is massive, the neural network needs to perform a time-consuming classification task. In this paper, a mechanical fault diagnosis method based on vibration signal down-sampling and the improved one-dimensional convolutional neural network is proposed. Through the robust principal component analysis, the low-rank feature matrix of a large amount of raw data can be separated, and then down-sampling is realized to reduce the subsequent calculation amount. In the improved one-dimensional CNN, a smaller convolution kernel is used to reduce the number of parameters and computational complexity, and regularization is introduced before the fully connected layer to prevent overfitting. In addition, the multi-connected layers can better generalize classification results without cumbersome parameter adjustments. The effectiveness of the method is verified by monitoring the signal of the centrifugal pump test bench, and the average test accuracy is above 98%. When compared with the traditional deep belief network (DBN) and support vector machine (SVM) methods, this method has better performance.

Keywords: fault diagnosis, vibration signal down-sampling, 1D-CNN

Procedia PDF Downloads 131

Authors: C. Brierley, H. El-Farahaty, A. Farhan

Abstract:

The Leeds Parallel Corpus of Arabic-English Constitutions is a parallel corpus for the Arabic legal domain. Analysis of legal language via Corpus Linguistics techniques is an important development. In legal proceedings, a corpus-based approach to disambiguating meaning is set to replace the dictionary as an interpretative tool, and legal scholarship in the States is now attuned to the potential for Text Analytics over vast quantities of text-based legal material, following the business and medical industries. This trend is reflected in Europe: the interdisciplinary research group in Computer Assisted Legal Linguistics mines big data collections of legal and non-legal texts to analyse: legal interpretations; legal discourse; the comprehensibility of legal texts; conflict resolution; and linguistic human rights. This paper focuses on ‘dignity’ as an important aspect of the overarching concept of human rights in current constitutions across the Arab world. We have compiled a parallel, Arabic-English raw text corpus (169,861 Arabic words and 205,893 English words) from reputable websites such as the World Intellectual Property Organisation and CONSTITUTE, and uploaded and queried our corpus in Sketch Engine. Our most challenging task was sentence-level alignment of Arabic-English data. This entailed manual intervention to ensure correspondence on a one-to-many basis since Arabic sentences differ from English in length and punctuation. We have searched for morphological variants of ‘dignity’ (رامة ك, karāma) in the Arabic data and inspected their English translation equivalents. The term occurs most frequently in the Sudanese constitution (10 instances), and not at all in the constitution of Palestine. Its most frequent collocate, determined via the logDice statistic in Sketch Engine, is ‘human’ as in ‘human dignity’.

Keywords: Arabic constitution, corpus-based legal linguistics, human rights, parallel Arabic-English legal corpora

Procedia PDF Downloads 183

Authors: Saeko Toyoshima

Abstract:

This research will discuss if performance tasks with technology would enhance students' willingness to communicate. The present study investigated how Japanese learners of English would change their attitude to communication in their target language by experiencing a performance task, called 'digital portfolio', in the classroom, applying the concepts of action research. The study adapted questionnaires including four-Likert and open-end questions as mixed-methods research. There were 28 students in the class. Many of Japanese university students with low proficiency (A1 in Common European Framework of References in Language Learning and Teaching) have difficulty in communicating in English due to the low proficiency and the lack of practice in and outside of the classroom at secondary education. They should need to mediate between themselves in the world of L1 and L2 with completing a performance task for communication. This paper will introduce the practice of CALL class where A1 level students have made their 'digital portfolio' related to the topics of TED® (Technology, Entertainment, Design) Talk materials. The students had 'Portfolio Session' twice in one term, once in the middle, and once at the end of the course, where they introduced their portfolio to their classmates and international students in English. The present study asked the students to answer a questionnaire about willingness to communicate twice, once at the end of the first term and once at the end of the second term. The four-Likert questions were statistically analyzed with a t-test, and the answers to open-end questions were analyzed to clarify the difference between them. They showed that the students had a more positive attitude to communication in English and enhanced their willingness to communicate through the experiences of the task. It will be the implication of this paper that making and presenting portfolio as a performance task would lead them to construct themselves in English and enable them to communicate with the others enjoyably and autonomously.

Keywords: action research, digital portfoliio, computer-assisted language learning, ELT with CALL system, mixed methods research, Japanese English learners, willingness to communicate

Procedia PDF Downloads 118

Authors: Strahinja Z. Kovačević, Lidija R. Jevrić, Sanja O. Podunavac Kuzmanović

Abstract:

The relationship between antibacterial activity of eighteen different substituted benzimidazole derivatives and their molecular characteristics was studied using chemometric QSAR (Quantitative Structure–Activity Relationships) approach. QSAR analysis has been carried out on inhibitory activity towards Staphylococcus aureus, by using molecular descriptors, as well as minimal inhibitory activity (MIC). Molecular descriptors were calculated from the optimized structures. Principal component analysis (PCA) followed by hierarchical cluster analysis (HCA) and multiple linear regression (MLR) was performed in order to select molecular descriptors that best describe the antibacterial behavior of the compounds investigated, and to determine the similarities between molecules. The HCA grouped the molecules in separated clusters which have the similar inhibitory activity. PCA showed very similar classification of molecules as the HCA, and displayed which descriptors contribute to that classification. MLR equations, that represent MIC as a function of the in silico molecular descriptors were established. The statistical significance of the estimated models was confirmed by standard statistical measures and cross-validation parameters (SD = 0.0816, F = 46.27, R = 0.9791, R2CV = 0.8266, R2adj = 0.9379, PRESS = 0.1116). These parameters indicate the possibility of application of the established chemometric models in prediction of the antibacterial behaviour of studied derivatives and structurally very similar compounds.

Keywords: antibacterial, benzimidazole, molecular descriptors, QSAR

Procedia PDF Downloads 364

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

Abstract:

Manufacturing companies are facing global competition and enormous cost pressure. The use of machine learning applications can help reduce production costs and create added value. Predictive quality enables the securing of product quality through data-supported predictions using machine learning models as a basis for decisions on test results. Furthermore, machine learning methods are able to process large amounts of data, deal with unfavourable row-column ratios and detect dependencies between the covariates and the given target as well as assess the multidimensional influence of all input variables on the target. Real production data are often subject to highly fluctuating boundary conditions and unbalanced data sets. Changes in production data manifest themselves in trends, systematic shifts, and seasonal effects. Thus, Machine learning applications require intensive pre-processing and feature selection. Data preprocessing includes rule-based data cleaning, the application of dimensionality reduction techniques, and the identification of comparable data subsets. Within the used real data set of Bosch hydraulic valves, the comparability of the same production conditions in the production of hydraulic valves within certain time periods can be identified by applying the concept drift method. Furthermore, a classification model is developed to evaluate the feature importance in different subsets within the identified time periods. By selecting comparable and stable features, the number of features used can be significantly reduced without a strong decrease in predictive power. The use of cross-process production data along the value chain of hydraulic valves is a promising approach to predict the quality characteristics of workpieces. In this research, the ada boosting classifier is used to predict the leakage of hydraulic valves based on geometric gauge blocks from machining, mating data from the assembly, and hydraulic measurement data from end-of-line testing. In addition, the most suitable methods are selected and accurate quality predictions are achieved.

Keywords: classification, achine learning, predictive quality, feature selection

Procedia PDF Downloads 162

Authors: Zhengkun Yi

Abstract:

Surface texture discrimination using artificial tactile sensors has attracted increasing attentions in the past decade as it can endow technical and robot systems with a key missing ability. However, as a major component of texture, roughness has rarely been explored. This paper presents an approach for tactile surface roughness discrimination, which includes two parts: (1) design and fabrication of a bioinspired artificial fingertip, and (2) tactile signal processing for tactile surface roughness discrimination. The bioinspired fingertip is comprised of two polydimethylsiloxane (PDMS) layers, a polymethyl methacrylate (PMMA) bar, and two perpendicular polyvinylidene difluoride (PVDF) film sensors. This artificial fingertip mimics human fingertips in three aspects: (1) Elastic properties of epidermis and dermis in human skin are replicated by the two PDMS layers with different stiffness, (2) The PMMA bar serves the role analogous to that of a bone, and (3) PVDF film sensors emulate Meissner’s corpuscles in terms of both location and response to the vibratory stimuli. Various extracted features and classification algorithms including support vector machines (SVM) and k-nearest neighbors (kNN) are examined for tactile surface roughness discrimination. Eight standard rough surfaces with roughness values (Ra) of 50 μm, 25 μm, 12.5 μm, 6.3 μm 3.2 μm, 1.6 μm, 0.8 μm, and 0.4 μm are explored. The highest classification accuracy of (82.6 ± 10.8) % can be achieved using solely one PVDF film sensor with kNN (k = 9) classifier and the standard deviation feature.

Keywords: bioinspired fingertip, classifier, feature extraction, roughness discrimination

Procedia PDF Downloads 312

Authors: Kaushik Sathupadi, Sandesh Achar

Abstract:

Human action recognition modeling is a critical task in machine learning. These systems require better techniques for recognizing body parts and selecting optimal features based on vision sensors to identify complex action patterns efficiently. Still, there is a considerable gap and challenges between images and videos, such as brightness, motion variation, and random clutters. This paper proposes a robust approach for classifying human actions over cloud-based image data. First, we apply pre-processing and detection, human and outer shape detection techniques. Next, we extract valuable information in terms of cues. We extract two distinct features: fuzzy local binary patterns and sequence representation. Then, we applied a greedy, randomized adaptive search procedure for data optimization and dimension reduction, and for classification, we used a random forest. We tested our model on two benchmark datasets, AAMAZ and the KTH Multi-view football datasets. Our HMR framework significantly outperforms the other state-of-the-art approaches and achieves a better recognition rate of 91% and 89.6% over the AAMAZ and KTH multi-view football datasets, respectively.

Keywords: computer vision, human motion analysis, random forest, machine learning

Procedia PDF Downloads 36

Authors: Eva Tvrdá, Boris Botman, Marek Halenár, Tomáš Slanina, Norbert Lukáč

Abstract:

In vitro storage and processing of animal semen represents a risk factor to spermatozoa vitality, potentially leading to reduced fertility. A variety of substances isolated from natural sources may exhibit protective or antioxidant properties on the spermatozoon, thus extending the lifespan of stored ejaculates. This study compared the ability of different concentrations of the Salvia officinalis extract on the motility, mitochondrial activity, viability and reactive oxygen species (ROS) production by bovine spermatozoa during different time periods (0, 2, 6 and 24 h) of in vitro culture. Spermatozoa motility was assessed using the Computer-assisted sperm analysis (CASA) system. Cell viability was examined using the metabolic activity MTT assay, the eosin-nigrosin staining technique was used to evaluate the sperm viability and ROS generation was quantified using luminometry. The CASA analysis revealed that the motility in the experimental groups supplemented with 0.5-2 µg/mL Salvia extract was significantly lower in comparison with the control (P<0.05; Time 24 h). At the same time, a long-term exposure of spermatozoa to concentrations ranging between 0.05 µg/mL and 2 µg/mL had a negative impact on the mitochondrial metabolism (P<0.05; Time 24 h). The viability staining revealed that 0.001-1 µg/mL Salvia extract had no effects on bovine male gametes, however 2 µg/mL Salvia had a persisting negative effect on spermatozoa (P<0.05). Furthermore 0.05-2 µg/mL Salvia exhibited an immediate ROS-promoting effect on the sperm culture (P>0.05; Time 0 h and 2 h), which remained significant throughout the entire in vitro culture (P<0.05; Time 24 h). Our results point out to the necessity to examine specific effects the biomolecules present in Salvia officinalis may have individually or collectively on the in vitro sperm vitality and oxidative profile.

Keywords: bulls, CASA, MTT test, reactive oxygen species, sage, Salvia officinalis, spermatozoa

Procedia PDF Downloads 338

Authors: Bandar Alahmadi, Lethia Jackson

Abstract:

Machine Learning model is used today in many real-life applications. The safety and security of such model is important, so the results of the model are as accurate as possible. One challenge of machine learning model security is the adversarial examples attack. Adversarial examples are designed by the attacker to cause the machine learning model to misclassify the input. We propose a method to generate adversarial examples to attack image classifiers. We are modifying the successfully classified images, so a classifier misclassifies them after the modification. In our method, we do not update the whole image, but instead we detect the important region, modify it, place it back to the original image, and then run it through a classifier. The algorithm modifies the detected region using two methods. First, it will add abstract image matrix on back of the detected image matrix. Then, it will perform a rotation attack to rotate the detected region around its axes, and embed the trace of image in image background. Finally, the attacked region is placed in its original position, from where it was removed, and a smoothing filter is applied to smooth the background with foreground. We test our method in cascade classifier, and the algorithm is efficient, the classifier confident has dropped to almost zero. We also try it in CNN (Convolutional neural network) with higher setting and the algorithm was successfully worked.

Keywords: adversarial examples, attack, computer vision, image processing

Procedia PDF Downloads 339

Authors: Tesfaye Mengistu

Abstract:

Power cables play a crucial role in the transmission and distribution of electrical energy. As the electricity generation, transmission, distribution, and storage systems become smarter, there is a growing emphasis on incorporating intelligent approaches to ensure the reliability of power cables. Various types of electrical cables are employed for transmitting and distributing electrical energy, with cross-linked polyethylene (XLPE) cables being widely utilized due to their exceptional electrical and mechanical properties. However, insulation defects can occur in XLPE cables due to subpar manufacturing techniques during production and cable joint installation. To address this issue, experts have proposed different methods for monitoring XLPE cables. Some suggest the use of interdigital capacitive (IDC) technology for online monitoring, while others propose employing continuous wave (CW) terahertz (THz) imaging systems to detect internal defects in XLPE plates used for power cable insulation. In this study, we have developed models that employ a custom dataset collected locally to classify the physical safety status of individual power cables. Our models aim to replace physical inspections with computer vision and image processing techniques to classify defective power cables from non-defective ones. The implementation of our project utilized the Python programming language along with the TensorFlow package and a convolutional neural network (CNN). The CNN-based algorithm was specifically chosen for power cable defect classification. The results of our project demonstrate the effectiveness of CNNs in accurately classifying power cable defects. We recommend the utilization of similar or additional datasets to further enhance and refine our models. Additionally, we believe that our models could be used to develop methodologies for detecting power cable defects from live video feeds. We firmly believe that our work makes a significant contribution to the field of power cable inspection and maintenance. Our models offer a more efficient and cost-effective approach to detecting power cable defects, thereby improving the reliability and safety of power grids.

Keywords: artificial intelligence, computer vision, defect detection, convolutional neural net

Procedia PDF Downloads 112

Authors: Kyomin Lee, Joohee Kim, Sangho Kang

Abstract:

The commercial nuclear power reactor in South Korea, Kori Unit 1, which was a 587 MWe pressurized water reactor that started operation since 1978, was permanently shut down in June 2017 without an additional operating license extension. The Kori 1 Unit is scheduled to become the nuclear power unit to enter the decommissioning phase. In this study, the preliminary evaluation of the decommissioning wastes for the Kori Unit 1 was performed based on the following series of process: firstly, the plant inventory is investigated based on various documents (i.e., equipment/ component list, construction records, general arrangement drawings). Secondly, the radiological conditions of systems, structures and components (SSCs) are established to estimate the amount of radioactive waste by waste classification. Third, the waste management strategies for Kori Unit 1 including waste packaging are established. Forth, selection of the proper decontamination and dismantling (D&D) technologies is made considering the various factors. Finally, the amount of decommissioning waste by classification for Kori 1 is estimated using the DeCAT program, which was developed by KEPCO-E&C for a decommissioning cost estimation. The preliminary evaluation results have shown that the expected amounts of decommissioning wastes were less than about 2% and 8% of the total wastes generated (i.e., sum of clean wastes and radwastes) before/after waste processing, respectively, and it was found that the majority of contaminated material was carbon or alloy steel and stainless steel. In addition, within the range of availability of information, the results of the evaluation were compared with the results from the various decommissioning experiences data or international/national decommissioning study. The comparison results have shown that the radioactive waste amount from Kori Unit 1 decommissioning were much less than those from the plants decommissioned in U.S. and were comparable to those from the plants in Europe. This result comes from the difference of disposal cost and clearance criteria (i.e., free release level) between U.S. and non-U.S. The preliminary evaluation performed using the methodology established in this study will be useful as a important information in establishing the decommissioning planning for the decommissioning schedule and waste management strategy establishment including the transportation, packaging, handling, and disposal of radioactive wastes.

Keywords: characterization, classification, decommissioning, decontamination and dismantling, Kori 1, radioactive waste

Procedia PDF Downloads 209

Authors: Kyi Pyar Zaw, Zin Mar Kyu

Abstract:

Character recognition is the process of converting a text image file into editable and searchable text file. Feature Extraction is the heart of any character recognition system. The character recognition rate may be low or high depending on the extracted features. In the proposed paper, 25 features for one character are used in character recognition. Basically, there are three steps of character recognition such as character segmentation, feature extraction and classification. In segmentation step, horizontal cropping method is used for line segmentation and vertical cropping method is used for character segmentation. In the Feature extraction step, features are extracted in two ways. The first way is that the 8 features are extracted from the entire input character using eight direction chain code frequency extraction. The second way is that the input character is divided into 16 blocks. For each block, although 8 feature values are obtained through eight-direction chain code frequency extraction method, we define the sum of these 8 feature values as a feature for one block. Therefore, 16 features are extracted from that 16 blocks in the second way. We use the number of holes feature to cluster the similar characters. We can recognize the almost Myanmar common characters with various font sizes by using these features. All these 25 features are used in both training part and testing part. In the classification step, the characters are classified by matching the all features of input character with already trained features of characters.

Keywords: chain code frequency, character recognition, feature extraction, features matching, segmentation

Procedia PDF Downloads 320

Authors: Mohammed Abdullah Alshahrani

Abstract:

The research focuses on investigating the use of partial least squares (PLS) methodology for addressing challenges associated with high-dimensional correlated data. Recent technological advancements have led to experiments producing data characterized by a large number of variables compared to observations, with substantial inter-variable correlations. Such data patterns are common in chemometrics, where near-infrared (NIR) spectrometer calibrations record chemical absorbance levels across hundreds of wavelengths, and in genomics, where thousands of genomic regions' copy number alterations (CNA) are recorded from cancer patients. PLS serves as a widely used method for analyzing high-dimensional data, functioning as a regression tool in chemometrics and a classification method in genomics. It handles data complexity by creating latent variables (components) from original variables. However, applying PLS can present challenges. The study investigates key areas to address these challenges, including unifying interpretations across three main PLS algorithms and exploring unusual negative shrinkage factors encountered during model fitting. The research presents an alternative approach to addressing the interpretation challenge of predictor weights associated with PLS. Sparse estimation of predictor weights is employed using a penalty function combining a lasso penalty for sparsity and a Cauchy distribution-based penalty to account for variable dependencies. The results demonstrate sparse and grouped weight estimates, aiding interpretation and prediction tasks in genomic data analysis. High-dimensional data scenarios, where predictors outnumber observations, are common in regression analysis applications. Ordinary least squares regression (OLS), the standard method, performs inadequately with high-dimensional and highly correlated data. Copy number alterations (CNA) in key genes have been linked to disease phenotypes, highlighting the importance of accurate classification of gene expression data in bioinformatics and biology using regularized methods like PLS for regression and classification.

Keywords: partial least square regression, genetics data, negative filter factors, high dimensional data, high correlated data

Procedia PDF Downloads 49