Search results for: post classification change detection

Authors: Nidhal K. Azawi, John M. Gauch

Abstract:

Colorectal cancer is one of the leading causes of cancer death in the US and the world, which is why millions of colonoscopy examinations are performed annually. Unfortunately, noise, specular highlights, and motion artifacts corrupt many images in a typical colonoscopy exam. The goal of our research is to produce automated techniques to detect and correct or remove these noninformative images from colonoscopy videos, so physicians can focus their attention on informative images. In this research, we first automatically extract features from images. Then we use machine learning and deep neural network to classify colonoscopy images as either informative or noninformative. Our results show that we achieve image classification accuracy between 92-98%. We also show how the removal of noninformative images together with image alignment can aid in the creation of image panoramas and other visualizations of colonoscopy images.

Keywords: colonoscopy classification, feature extraction, image alignment, machine learning

Procedia PDF Downloads 253

Authors: Faisal Aburub, Wael Hadi

Abstract:

Data mining is the process of extracting useful or hidden information from a large database. Extracted information can be used to discover relationships among features, where data objects are grouped according to logical relationships; or to predict unseen objects to one of the predefined groups. In this paper, we aim to investigate four well-known data mining algorithms in order to predict groundwater areas in Jordan. These algorithms are Support Vector Machines (SVMs), Naïve Bayes (NB), K-Nearest Neighbor (kNN) and Classification Based on Association Rule (CBA). The experimental results indicate that the SVMs algorithm outperformed other algorithms in terms of classification accuracy, precision and F1 evaluation measures using the datasets of groundwater areas that were collected from Jordanian Ministry of Water and Irrigation.

Keywords: classification, data mining, evaluation measures, groundwater

Procedia PDF Downloads 281

Authors: Kuei-Ling Sun, Emily Chia-Yu Su

Abstract:

Allergy is a hypersensitive overreaction of the immune system to environmental stimuli, and a major health problem. These overreactions include rashes, sneezing, fever, food allergies, anaphylaxis, asthmatic, shock, or other abnormal conditions. Allergies can be caused by food, insect stings, pollen, animal wool, and other allergens. Their development of allergies is due to both genetic and environmental factors. Allergies involve immunoglobulin E antibodies, a part of the body’s immune system. Immunoglobulin E antibodies will bind to an allergen and then transfer to a receptor on mast cells or basophils triggering the release of inflammatory chemicals such as histamine. Based on the increasingly serious problem of environmental change, changes in lifestyle, air pollution problem, and other factors, in this study, we both collect allergens and non-allergens from several databases and use several machine learning methods for classification, including logistic regression (LR), stepwise regression, decision tree (DT) and neural networks (NN) to do the model comparison and determine the permutations of allergen amino acid’s sequence.

Keywords: allergy, classification, decision tree, logistic regression, machine learning

Procedia PDF Downloads 305

Authors: Mouna Zitouni, Mariem Tounsi

Abstract:

This research introduces an application of the Riesz mixture model for medical image segmentation for accurate diagnosis and treatment of brain tumors. We propose a pixel classification technique based on the Riesz distribution, derived from an extended Bartlett decomposition. To our knowledge, this is the first study addressing this approach. The Expectation-Maximization algorithm is implemented for parameter estimation. A comparative analysis, using both synthetic and real brain images, demonstrates the superiority of the Riesz model over a recent method based on the Wishart distribution.

Keywords: EM algorithm, segmentation, Riesz probability distribution, Wishart probability distribution

Procedia PDF Downloads 21

Authors: Qiaowen Lin

Abstract:

The 3D cadastre is an inevitable choice to meet the needs of real cadastral management. Nowadays, more attention is given to the technical aspects of 3D cadastre, resulting in the imbalance within this field. To fulfill this research gap, the stakeholder, which has been regarded as the determining factor in cadastral change has been studied. Delphi method, Michael rating, and stakeholder mapping are used to identify and classify the stakeholders in 3D cadastre. It is concluded that the project managers should pay more attention to the interesting appeal of the key stakeholders and different coping strategies should be adopted to facilitate the transition to 3D cadastre.

Keywords: stakeholders, three dimension, cadastre, transtion

Procedia PDF Downloads 290

Authors: D. I. George Amalarethinam, A. Emima

Abstract:

Educational Data Mining (EDM) is one of the newest topics to emerge in recent years, and it is concerned with developing methods for analyzing various types of data gathered from the educational circle. EDM methods and techniques with machine learning algorithms are used to extract meaningful and usable information from huge databases. For scientists and researchers, realistic applications of Machine Learning in the EDM sectors offer new frontiers and present new problems. One of the most important research areas in EDM is predicting student success. The prediction algorithms and techniques must be developed to forecast students' performance, which aids the tutor, institution to boost the level of student’s performance. This paper examines various classification techniques in prediction methods and data mining tools used in EDM.

Keywords: classification technique, data mining, EDM methods, prediction methods

Procedia PDF Downloads 118

Authors: Hanaa M. Abdelgawad, Mona Safar, Ayman M. Wahba

Abstract:

Real-time image and video processing is a demand in many computer vision applications, e.g. video surveillance, traffic management and medical imaging. The processing of those video applications requires high computational power. Therefore, the optimal solution is the collaboration of CPU and hardware accelerators. In this paper, a Canny edge detection hardware accelerator is proposed. Canny edge detection is one of the common blocks in the pre-processing phase of image and video processing pipeline. Our presented approach targets offloading the Canny edge detection algorithm from processing system (PS) to programmable logic (PL) taking the advantage of High Level Synthesis (HLS) tool flow to accelerate the implementation on Zynq platform. The resulting implementation enables up to a 100x performance improvement through hardware acceleration. The CPU utilization drops down and the frame rate jumps to 60 fps of 1080p full HD input video stream.

Keywords: high level synthesis, canny edge detection, hardware accelerators, computer vision

Procedia PDF Downloads 480

Authors: Jaspreet Singh, Gurvinder Singh, Prabhsimran Singh, Rajinder Singh, Prithvipal Singh, Karanjeet Singh Kahlon, Ravinder Singh Sawhney

Abstract:

Morphological evaluation of Indian languages is one of the burgeoning fields in the area of Natural Language Processing (NLP). The evaluation of a language is an eminent task in the era of information retrieval and text mining. The extraction and classification of knowledge from text can be exploited for sentiment analysis and morphological evaluation. This study coalesce morphological evaluation and sentiment analysis for the task of classification of farmer suicide cases reported in Punjab state of India. The pre-processing of Punjabi text involves morphological evaluation and normalization of Punjabi word tokens followed by the training of proposed model using deep learning classification on Punjabi language text extracted from online Punjabi news reports. The class-wise accuracies of sentiment prediction for four negatively oriented classes of farmer suicide cases are 93.85%, 88.53%, 83.3%, and 95.45% respectively. The overall accuracy of sentiment classification obtained using proposed framework on 275 Punjabi text documents is found to be 90.29%.

Keywords: deep neural network, farmer suicides, morphological processing, punjabi text, sentiment analysis

Procedia PDF Downloads 328

Authors: Pei-Jyun Hsieh, Cheng-Hsuan Li, Bor-Chen Kuo

Abstract:

For hyperspectral image classification, feature reduction is an important pre-processing for avoiding the Hughes phenomena due to the difficulty for collecting training samples. Hence, lots of researches developed feature selection methods such as F-score, HSIC (Hilbert-Schmidt Independence Criterion), and etc., to improve hyperspectral image classification. However, most of them only consider the class separability in the original space, i.e., a linear class separability. In this study, we proposed a nonlinear class separability measure based on kernel trick for selecting an appropriate feature subset. The proposed nonlinear class separability was formed by a generalized RBF kernel with different bandwidths with respect to different features. Moreover, it considered the within-class separability and the between-class separability. A genetic algorithm was applied to tune these bandwidths such that the smallest with-class separability and the largest between-class separability simultaneously. This indicates the corresponding feature space is more suitable for classification. In addition, the corresponding nonlinear classification boundary can separate classes very well. These optimal bandwidths also show the importance of bands for hyperspectral image classification. The reciprocals of these bandwidths can be viewed as weights of bands. The smaller bandwidth, the larger weight of the band, and the more importance for classification. Hence, the descending order of the reciprocals of the bands gives an order for selecting the appropriate feature subsets. In the experiments, three hyperspectral image data sets, the Indian Pine Site data set, the PAVIA data set, and the Salinas A data set, were used to demonstrate the selected feature subsets by the proposed nonlinear feature selection method are more appropriate for hyperspectral image classification. Only ten percent of samples were randomly selected to form the training dataset. All non-background samples were used to form the testing dataset. The support vector machine was applied to classify these testing samples based on selected feature subsets. According to the experiments on the Indian Pine Site data set with 220 bands, the highest accuracies by applying the proposed method, F-score, and HSIC are 0.8795, 0.8795, and 0.87404, respectively. However, the proposed method selects 158 features. F-score and HSIC select 168 features and 217 features, respectively. Moreover, the classification accuracies increase dramatically only using first few features. The classification accuracies with respect to feature subsets of 10 features, 20 features, 50 features, and 110 features are 0.69587, 0.7348, 0.79217, and 0.84164, respectively. Furthermore, only using half selected features (110 features) of the proposed method, the corresponding classification accuracy (0.84168) is approximate to the highest classification accuracy, 0.8795. For other two hyperspectral image data sets, the PAVIA data set and Salinas A data set, we can obtain the similar results. These results illustrate our proposed method can efficiently find feature subsets to improve hyperspectral image classification. One can apply the proposed method to determine the suitable feature subset first according to specific purposes. Then researchers can only use the corresponding sensors to obtain the hyperspectral image and classify the samples. This can not only improve the classification performance but also reduce the cost for obtaining hyperspectral images.

Keywords: hyperspectral image classification, nonlinear feature selection, kernel trick, support vector machine

Procedia PDF Downloads 265

Authors: Shunzuo Wu, Xudong Luo, Yuanxiu Liao

Abstract:

Recently, the rapid development of deep learning makes artificial intelligence (AI) penetrate into many fields, replacing manual work there. In particular, AI systems also become a research focus in the field of automatic office. To meet real needs in automatic officiating, in this paper we develop an automatic form filling system. Specifically, it uses two classical neural network models and several word embedding models to classify various relevant information elicited from the Internet. When training the neural network models, we use less noisy and balanced data for training. We conduct a series of experiments to test my systems and the results show that our system can achieve better classification results.

Keywords: artificial intelligence and office, NLP, deep learning, text classification

Procedia PDF Downloads 202

Authors: Yu Guan, Song Lu, Wei Yuan, Heping Zhang

Abstract:

Gas fire extinguishing system is widely used due to its cleanliness and efficiency, and since its spray will be affected by many factors such as convection and obstacles in jetting region, so in order to evaluate its effectiveness, detecting concentration distribution in the jetting area is indispensable, which is commonly achieved by aspirating concentration detection technique. During the concentration measurement, the response time of detector is a very important parameter, especially for those fire-extinguishing systems with rapid gas dispersion. Long response time will not only underestimate its concentration but also prolong the change of concentration with time. Therefore it is necessary to analyze the factors influencing the response time. In the paper, an aspirating concentration detection method was introduced, which is achieved by using a small critical nozzle and a laminar flowmeter, and because of the response time is mainly related to the gas transport process from sampling site to the sensor, the effects of exhaust pipe size, gas flow rate, and gas concentration on its response time were analyzed. During the research, Bromotrifluoromethane (CBrF₃) was used. The effect of the sampling tube was investigated with different length of 1, 2, 3, 4 and 5 m (5mm in pipe diameter) and different pipe diameter of 3, 4, 5, 6 and 8 mm (3m in length). The effect of gas flow rate was analyzed by changing the throat diameter of the critical nozzle with 0.5, 0.682, 0.75, 0.8, 0.84 and 0.88 mm. The effect of gas concentration on response time was studied with the concentration range of 0-25%. The result showed that the response time increased with the increase of both the length and diameter of the sampling pipe, and the effect of length on response time was linear, but for the effect of diameter, it was exponential. It was also found that as the throat diameter of critical nozzle increased, the response time reduced a lot, in other words, gas flow rate has a great influence on response time. For the effect of gas concentration, the response time increased with the increase of the CBrF₃ concentration, and the slope of the curve was reduced.

Keywords: aspirating concentration detection, fire extinguishing, gaseous agent, response time

Procedia PDF Downloads 271

Authors: Manoj Parmar

Abstract:

The article uses reflexivity as a research method to investigate and propose an architectural theory plan for climate change. It hypothecates that to discuss or formulate discourse on "Architectural Thinking in a Time of Climate Emergency," firstly, we need to understand the modes of integration that enable architectural thinking with climate change. The study intends to study the various integration modes that have evolved historically and situate them in time. Subsequently, it analyses the integration pattern, challenges the existing model, and finds a way towards climate change as central to architectural thinking. The study is fundamental on-premises that ecology and climate change scholarship has consistently out lashed the asymmetrical and nonlinear knowledge and needs approaches for architecture that are less burden to climate change to people and minimize its impact on ecology.

Keywords: climate change, architectural theory, reflexivity, modernity

Procedia PDF Downloads 285

Authors: Ahmadu Maidorawa, Idrissa Djibo, Muhammad Tella

Abstract:

This paper presents a method for detecting flow chart and figure plagiarism based on shape of image processing and multimedia retrieval. The method managed to retrieve flowcharts with ranked similarity according to different matching sets. Plagiarism detection is well known phenomenon in the academic arena. Copying other people is considered as serious offense that needs to be checked. There are many plagiarism detection systems such as turn-it-in that has been developed to provide these checks. Most, if not all, discard the figures and charts before checking for plagiarism. Discarding the figures and charts result in look holes that people can take advantage. That means people can plagiarize figures and charts easily without the current plagiarism systems detecting it. There are very few papers which talks about flowcharts plagiarism detection. Therefore, there is a need to develop a system that will detect plagiarism in figures and charts.

Keywords: flowchart, multimedia retrieval, figures similarity, image comparison, figure retrieval

Procedia PDF Downloads 467

Authors: Ahmad H. Abdelgwad, Ahmed A. Nashat

Abstract:

Landmine detection is an important and yet challenging problem remains to be solved. Ground Penetrating Radar (GPR) is a powerful and rapidly maturing technology for subsurface threat identification. The detection methodology of GPR depends mainly on the contrast of the dielectric properties of the searched target and its surrounding soil. This contrast produces a partial reflection of the electromagnetic pulses that are being transmitted into the soil and then being collected by the GPR.  One of the most critical hardware components for the performance of GPR is the antenna system. The current paper explores the design and simulation of a pyramidal horn antenna operating at L-band frequencies (1- 2 GHz) to detect a landmine. A prototype model of the GPR system setup is developed to simulate full wave analysis of the electromagnetic fields in different soil types. The contrast in the dielectric permittivity of the landmine and the sandy soil is the most important parameter to be considered for detecting the presence of landmine. L-band horn antenna is proved to be well-versed in the investigation of landmine detection.

Keywords: full wave analysis, ground penetrating radar, horn antenna design, landmine detection

Procedia PDF Downloads 221

Authors: Masood Abdollahi, Seyed Naser Moghaddas Tafreshi

Abstract:

Expanded polystyrene (EPS) geofoam is a cellular geosynthetic material that can be used to protect lifelines (e.g. pipelines, electricity cables, etc.) below ground. Post and beam system is the most recent configuration of EPS blocks which can be implemented for this purpose. It provides a void space atop lifelines which allows settlement of the loading surface with imposing no pressure on the lifelines system. This paper investigates the efficiency of the configuration of post-beam system subjected to static loading. To evaluate the soil surface settlement, beam deformation and transferred pressure over the beam, laboratory tests using two different densities for EPS blocks are conducted. The effect of geogrid-reinforcing the cover soil on system response is also investigated. The experimental results show favorable performance of EPS post and beam configuration in protecting underground lifelines. 

Keywords: beam deformation, EPS block, laboratory test, post-Beam system, soil surface settlement

Procedia PDF Downloads 238

Authors: Rodrigo Costa, Thomas Mazzuchi, Shahram Sarkani

Abstract:

This study analyzes the evolution of Bitcoin Improvement Proposals (BIPs), the self-governing mechanism that enables updates to the Bitcoin protocol. By modeling BIP submission frequencies with a Negative Binomial distribution and detecting change points with the Pelt Rupture model, we identify three distinct intervals of proposal activity, suggesting shifts in development priorities over time. Long-term growth patterns, captured by Gompertz and Weibull models, indicate an S-shaped trend in cumulative BIP counts, pointing toward a maturation phase in Bitcoin’s protocol. Our findings suggest that Bitcoin may be entering a stable stage, with fewer fundamental changes and more incremental enhancements. This trend highlights the need for further research into BIP content and more studies into its dynamics to better understand decentralized protocol governance and maturation.

Keywords: bitcoin improvement proposals, innovation management, change point detection, systems modeling, simulation

Procedia PDF Downloads 19

Authors: David Idiata

Abstract:

This paper is directed at using the UKCP09 climate change projection tool to predict the effect of climate change on the dynamic modulus of asphalt concrete is Southern England knowing that there is a pressing challenge directly facing infrastructure in the urban cities in the world today due to climate change. Climate change causes change in the environment which in turn impacts on the long-term structural performance of structures. From the projection values obtained, it was discovered that as the temperature increases, the dynamic modulus reduces and this effect was more on the South West which have temperature range of 36.8 oC to 48.3 oC and dynamic modulus range of 2,212 MPa to 1256 MPa.

Keywords: dynamic modulus, asphalt concrete, UKCP09, Southern England

Procedia PDF Downloads 360

Authors: Liu Linxin

Abstract:

As system complexity increases, log parsing and anomaly detection become more and more important in ensuring system stability. However, traditional methods often face the problems of insufficient adaptability and decreasing accuracy when dealing with rapidly changing log contents and unknown domains. To this end, this paper proposes an approach LogRAG, which combines RAG (Retrieval Augmented Generation) technology with Prompt Engineering for Large Language Models, applied to log analysis tasks to achieve dynamic parsing of logs and intelligent anomaly detection. By combining real-time information retrieval and prompt optimisation, this study significantly improves the adaptive capability of log analysis and the interpretability of results. Experimental results show that the method performs well on several public datasets, especially in the absence of training data, and significantly outperforms traditional methods. This paper provides a technical path for log parsing and anomaly detection, demonstrating significant theoretical value and application potential.

Keywords: log parsing, anomaly detection, retrieval-augmented generation, prompt engineering, LLMs

Procedia PDF Downloads 31

Authors: Maureen Royce, Joshi Jariwala, Sally Kah

Abstract:

Change is inevitable, but the change process is progressive. Organisational change is the process in which an organisation changes strategies, operational methods, systems, culture, and structure to affect something different in the organisation. This process can be continuous or developed over a period and driven by internal and external factors. Organisational change is essential if organisations are to survive in dynamic and uncertain environments. However, evidence from research shows that many change initiatives fail, leading to severe consequences for organisations and their resources. The complex models of third sector organisations, i.e., social enterprise, compounds the levels of change in these organisations. Interestingly, innovation is associated with a change in social enterprises due to the hybridity of product and service development. Furthermore, the creation of social intervention has offered a new process and outcomes to the lifecycle of change. Therefore, different forms of organisational innovation are developed, i.e., total, evolutionary, expansionary, and developmental, which affect the interventions of social enterprises. This raises both theoretical and business concerns on how the competing hybrid nature of social enterprises change, how change is managed, and the impact on these organisations. These perspectives present critical questions for further investigation. In this study, we investigate the impact of organisational change on employees and organisational development at DaDaFest –a disability arts organisation with a social focus based in Liverpool. The three main objectives are to explore the drivers of change and the implementation process; to examine the impact of organisational change on employees and; to identify barriers to organisation change and development. To address the preceding research objectives, qualitative research design is adopted using semi-structured interviews. Data is analysed using a six-step thematic analysis framework, which enables the study to develop themes depicting the impact of change on employees and organisational development. This study presents theoretical and practical contributions for academics and practitioners. The knowledge contributions encapsulate the evolution of change and the change cycle in a social enterprise. However, practical implications provide critical insights into the change management process and the impact of change on employees and organisational development.

Keywords: organisational change, change management, organisational change system, social enterprise

Procedia PDF Downloads 128

Authors: Bingchun Liu, Pei-Chann Chang, Natasha Huang, Dun Li

Abstract:

Machine Learning and Data Mining are the two important tools for extracting useful information and knowledge from large datasets. In machine learning, classification is a wildly used technique to predict qualitative variables and is generally preferred over regression from an operational point of view. Due to the enormous increase in air pollution in various countries especially China, Air Quality Classification has become one of the most important topics in air quality research and modelling. This study aims at introducing a hybrid classification model based on information theory and Support Vector Machine (SVM) using the air quality data of four cities in China namely Beijing, Guangzhou, Shanghai and Tianjin from Jan 1, 2014 to April 30, 2016. China's Ministry of Environmental Protection has classified the daily air quality into 6 levels namely Serious Pollution, Severe Pollution, Moderate Pollution, Light Pollution, Good and Excellent based on their respective Air Quality Index (AQI) values. Using the information theory, information gain (IG) is calculated and feature selection is done for both categorical features and continuous numeric features. Then SVM Machine Learning algorithm is implemented on the selected features with cross-validation. The final evaluation reveals that the IG and SVM hybrid model performs better than SVM (alone), Artificial Neural Network (ANN) and K-Nearest Neighbours (KNN) models in terms of accuracy as well as complexity.

Keywords: machine learning, air quality classification, air quality index, information gain, support vector machine, cross-validation

Procedia PDF Downloads 236

Authors: Belkacem Soundes, Guezouli Larbi

Abstract:

This paper presents a robust and accurate method for text detection and localization over lecture videos. Frame regions are classified into text or background based on visual feature analysis. However, lecture video shows significant degradation mainly related to acquisition conditions, camera motion and environmental changes resulting in low quality videos. Hence, affecting feature extraction and description efficiency. Moreover, traditional text detection methods cannot be directly applied to lecture videos. Therefore, robust feature extraction methods dedicated to this specific video genre are required for robust and accurate text detection and extraction. Method consists of a three-step process: Slide region detection and segmentation; Feature extraction and non-text filtering. For robust and effective features extraction moment functions are used. Two distinct types of moments are used: orthogonal and non-orthogonal. For orthogonal Zernike Moments, both Pseudo Zernike moments are used, whereas for non-orthogonal ones Hu moments are used. Expressivity and description efficiency are given and discussed. Proposed approach shows that in general, orthogonal moments show high accuracy in comparison to the non-orthogonal one. Pseudo Zernike moments are more effective than Zernike with better computation time.

Keywords: text detection, text localization, lecture videos, pseudo zernike moments

Procedia PDF Downloads 153

Authors: Moschos Vogiatzis, K. Perakis

Abstract:

Spatial-explicit and up-to-date land use/land cover information is fundamental for spatial planning, land management, sustainable development, and sound decision-making. In the last decade, many satellite-derived land cover products at different spatial, spectral, and temporal resolutions have been developed, such as the European Copernicus Land Cover product. However, more efficient and detailed information for land use/land cover is required at the regional or local scale. A typical Mediterranean basin with a complex landscape comprised of various forest types, crops, artificial surfaces, and wetlands was selected to test and develop our approach. In this study, we investigate the improvement of Copernicus Land Cover product (CLC2018) using Landsat 8 and Sentinel-2 pixel-based classification based on all available existing geospatial data (Forest Maps, LPIS, Natura2000 habitats, cadastral parcels, etc.). We examined and compared the performance of the Random Forest classifier for land use/land cover mapping. In total, 10 land use/land cover categories were recognized in Landsat 8 and 11 in Sentinel-2A. A comparison of the overall classification accuracies for 2018 shows that Landsat 8 classification accuracy was slightly higher than Sentinel-2A (82,99% vs. 80,30%). We concluded that the main land use/land cover types of CLC2018, even within a heterogeneous area, can be successfully mapped and updated according to CLC nomenclature. Future research should be oriented toward integrating spatiotemporal information from seasonal bands and spectral indexes in the classification process.

Keywords: classification, land use/land cover, mapping, random forest

Procedia PDF Downloads 127

Authors: Noura Al Akkari, Aurélie Foucquier, Sylvain Lespinats

Abstract:

Load monitoring is a management process for energy consumption towards energy savings and energy efficiency. Non Intrusive Load Monitoring (NILM) is one method of load monitoring used for disaggregation purposes. NILM is a technique for identifying individual appliances based on the analysis of the whole residence data retrieved from the main power meter of the house. Our NILM framework starts with data acquisition, followed by data preprocessing, then event detection, feature extraction, then general appliance modeling and identification at the final stage. The event detection stage is a core component of NILM process since event detection techniques lead to the extraction of appliance features. Appliance features are required for the accurate identification of the household devices. In this research work, we aim at developing a new event detection methodology with accurate load disaggregation to extract appliance features. Time-domain features extracted are used for tuning general appliance models for appliance identification and classification steps. We use unsupervised algorithms such as Dynamic Time Warping (DTW). The proposed method relies on detecting areas of operation of each residential appliance based on the power demand. Then, detecting the time at which each selected appliance changes its states. In order to fit with practical existing smart meters capabilities, we work on low sampling data with a frequency of (1/60) Hz. The data is simulated on Load Profile Generator software (LPG), which was not previously taken into consideration for NILM purposes in the literature. LPG is a numerical software that uses behaviour simulation of people inside the house to generate residential energy consumption data. The proposed event detection method targets low consumption loads that are difficult to detect. Also, it facilitates the extraction of specific features used for general appliance modeling. In addition to this, the identification process includes unsupervised techniques such as DTW. To our best knowledge, there exist few unsupervised techniques employed with low sampling data in comparison to the many supervised techniques used for such cases. We extract a power interval at which falls the operation of the selected appliance along with a time vector for the values delimiting the state transitions of the appliance. After this, appliance signatures are formed from extracted power, geometrical and statistical features. Afterwards, those formed signatures are used to tune general model types for appliances identification using unsupervised algorithms. This method is evaluated using both simulated data on LPG and real-time Reference Energy Disaggregation Dataset (REDD). For that, we compute performance metrics using confusion matrix based metrics, considering accuracy, precision, recall and error-rate. The performance analysis of our methodology is then compared with other detection techniques previously used in the literature review, such as detection techniques based on statistical variations and abrupt changes (Variance Sliding Window and Cumulative Sum).

Keywords: electrical disaggregation, DTW, general appliance modeling, event detection

Procedia PDF Downloads 78

Authors: Bernd Kiefer, Abraham Gebru Tesfay, Dietrich Klakow

Abstract:

The motivation of our work is to detect different terrain types traversed by a robot based on acoustic data from the robot-terrain interaction. Different acoustic features and classifiers were investigated, such as Mel-frequency cepstral coefficient and Gamma-tone frequency cepstral coefficient for the feature extraction, and Gaussian mixture model and Feed forward neural network for the classification. We analyze the system’s performance by comparing our proposed techniques with some other features surveyed from distinct related works. We achieve precision and recall values between 87% and 100% per class, and an average accuracy at 95.2%. We also study the effect of varying audio chunk size in the application phase of the models and find only a mild impact on performance.

Keywords: acoustic features, autonomous robots, feature extraction, terrain classification

Procedia PDF Downloads 370

Authors: Mohamed R. Mhereeg

Abstract:

The paper discusses the implementation of the MultiAgent classification System (MACS) and utilizing it to provide an automated and accurate classification of end users developing applications in the spreadsheet domain. However, different technologies have been brought together to build MACS. The strength of the system is the integration of the agent technology with the FIPA specifications together with other technologies, which are the .NET widows service based agents, the Windows Communication Foundation (WCF) services, the Service Oriented Architecture (SOA), and Oracle Data Mining (ODM). Microsoft's .NET windows service based agents were utilized to develop the monitoring agents of MACS, the .NET WCF services together with SOA approach allowed the distribution and communication between agents over the WWW. The Monitoring Agents (MAs) were configured to execute automatically to monitor excel spreadsheets development activities by content. Data gathered by the Monitoring Agents from various resources over a period of time was collected and filtered by a Database Updater Agent (DUA) residing in the .NET client application of the system. This agent then transfers and stores the data in Oracle server database via Oracle stored procedures for further processing that leads to the classification of the end user developers.

Keywords: MACS, implementation, multi-agent, SOA, autonomous, WCF

Procedia PDF Downloads 275

Authors: Bandar Alahmadi, Manohar Mareboyana, Lethia Jackson

Abstract:

Today, there are many applications that are using computer vision models, such as face recognition, image classification, and object detection. The accuracy of these models is very important for the performance of these applications. One challenge that facing the computer vision models is the adversarial examples attack. In computer vision, the adversarial example is an image that is intentionally designed to cause the machine learning model to misclassify it. One of very well-known method that is used to attack the Convolution Neural Network (CNN) is Fast Gradient Sign Method (FGSM). The goal of this method is to find the perturbation that can fool the CNN using the gradient of the cost function of CNN. In this paper, we introduce a novel model that can attack Regional-Convolution Neural Network (R-CNN) that use FGSM. We first extract the regions that are detected by R-CNN, and then we resize these regions into the size of regular images. Then, we find the best perturbation of the regions that can fool CNN using FGSM. Next, we add the resulted perturbation to the attacked region to get a new region image that looks similar to the original image to human eyes. Finally, we placed the regions back to the original image and test the R-CNN with the attacked images. Our model could drop the accuracy of the R-CNN when we tested with Pascal VOC 2012 dataset.

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

Procedia PDF Downloads 193

Authors: Yushen Dai, Tao Deng, Miaoying Chen, Baoqin Huang, Yan Ji, Yongshen Feng, Shaofei Liu, Dongmei Zhong, Tao Zhang, Lifeng Zhang

Abstract:

Background: Detection and diagnosis are prerequisites for early interventions in the care of children with Autism Spectrum Disorder (ASD). However, few studies have focused on this topic. Aim: This study aims to characterize the timing from symptom detection to intervention in children with ASD and to identify the potential predictors of early detection, diagnosis, and intervention. Methods and procedures: A cross-sectional survey was conducted with 314 parents of children with ASD in Guangzhou, China. Outcomes and Results: This study found that most children (76.24%) were diagnosed within one year after detection, and 25.8% of them did not receive the intervention after diagnosis. Predictors to ASD diagnosis included ASD-related symptoms identified at a younger age, more serious symptoms, and initial symptoms with abnormal development and sensory anomalies. ASD-related symptoms observed at an older age, initial symptoms with the social deficit, sensory anomalies, and without language impairment, parents as the primary caregivers, family with lower income and less social support utilization increased the odds of the time lag between detection and diagnosis. Children whose fathers had a lower level of education were less likely to receive the intervention. Conclusions and Implications: The study described the time for detection, diagnosis, and interventions of children with ASD. Findings suggest that the ASD-related symptoms, the timing at which symptoms first become a concern, primary caregivers’ roles, father’s educational level, and the family economic status should be considered when offering support to improve early detection, diagnosis, and intervention. Helping children and their families take full advantage of support is also important.

Keywords: autism spectrum disorder, child, detection, diagnosis, intervention, social support

Procedia PDF Downloads 93

Authors: Rim Messaoudi, Nogaye-Gueye Gning, François Azelart

Abstract:

Automatic text classification applies mostly natural language processing (NLP) and other AI-guided techniques to automatically classify text in a faster and more accurate manner. This paper discusses the subject of using predictive maintenance to manage incident tickets inside the sociality. It focuses on proposing a tool that treats and analyses comments and notes written by administrators after resolving an incident ticket. The goal here is to increase the quality of these comments. Additionally, this tool is based on NLP and machine learning techniques to realize the textual analytics of the extracted data. This approach was tested using real data taken from the French National Railways (SNCF) company and was given a high-quality result.

Keywords: machine learning, text classification, NLP techniques, semantic representation

Procedia PDF Downloads 103

Authors: Gizelle K. Vianna, Gabriel V. Cunha, Gustavo S. Oliveira

Abstract:

Foliage diseases in plants can cause a reduction in both quality and quantity of agricultural production. Intelligent detection of plant diseases is an essential research topic as it may help monitoring large fields of crops by automatically detecting the symptoms of foliage diseases. This work investigates ways to recognize the late blight disease from the analysis of tomato digital images, collected directly from the field. A pair of multilayer perceptron neural network analyzes the digital images, using data from both RGB and HSL color models, and classifies each image pixel. One neural network is responsible for the identification of healthy regions of the tomato leaf, while the other identifies the injured regions. The outputs of both networks are combined to generate the final classification of each pixel from the image and the pixel classes are used to repaint the original tomato images by using a color representation that highlights the injuries on the plant. The new images will have only green, red or black pixels, if they came from healthy or injured portions of the leaf, or from the background of the image, respectively. The system presented an accuracy of 97% in detection and estimation of the level of damage on the tomato leaves caused by late blight.

Keywords: artificial neural networks, digital image processing, pattern recognition, phytosanitary

Procedia PDF Downloads 330

Authors: Mohammed Arif, Abdessalam Kifouche

Abstract:

Optical character recognition (OCR) has a main role in the present time. It's capable to solve many serious problems and simplify human activities. The OCR yields to 70's, since many solutions has been proposed, but unfortunately, it was supportive to nothing but Latin languages. This work proposes a system of recognition of an off-line Arabic handwriting. This system is based on a structural segmentation method and uses support vector machines (SVM) in the classification phase. We have presented a state of art of the characters segmentation methods, after that a view of the OCR area, also we will address the normalization problems we went through. After a comparison between the Arabic handwritten characters & the segmentation methods, we had introduced a contribution through a segmentation algorithm.

Keywords: OCR, segmentation, Arabic characters, PAW, post-processing, SVM

Procedia PDF Downloads 74