Search results for: feature method

Authors: Zhang Shuqi, Liu Dan

Abstract:

For the problems in intelligent network anomaly traffic detection models, such as low detection accuracy caused by the lack of training samples, poor effect with small sample attack detection, a classification model enhancement method, F-ACGAN(Flow Auxiliary Classifier Generative Adversarial Network) which introduces generative adversarial network and adversarial training, is proposed to solve these problems. Generating adversarial data with category labels could enhance the training effect and improve classification accuracy and model robustness. FACGAN consists of three steps: feature preprocess, which includes data type conversion, dimensionality reduction and normalization, etc.; A generative adversarial network model with feature learning ability is designed, and the sample generation effect of the model is improved through adversarial iterations between generator and discriminator. The adversarial disturbance factor of the gradient direction of the classification model is added to improve the diversity and antagonism of generated data and to promote the model to learn from adversarial classification features. The experiment of constructing a classification model with the UNSW-NB15 dataset shows that with the enhancement of FACGAN on the basic model, the classification accuracy has improved by 8.09%, and the score of F1 has improved by 6.94%.

Keywords: data imbalance, GAN, ACGAN, anomaly detection, adversarial training, data augmentation

Procedia PDF Downloads 104

Authors: Ahmet Karagoz, Irfan Karagoz

Abstract:

Synthetic Aperture Radar (SAR) is a radar mechanism that can be integrated into manned and unmanned aerial vehicles to create high-resolution images in all weather conditions, regardless of day and night. In this study, SAR images of military vehicles with different azimuth and descent angles are pre-processed at the first stage. The main purpose here is to reduce the high speckle noise found in SAR images. For this, the Wiener adaptive filter, the mean filter, and the median filters are used to reduce the amount of speckle noise in the images without causing loss of data. During the image segmentation phase, pixel values are ordered so that the target vehicle region is separated from other regions containing unnecessary information. The target image is parsed with the brightest 20% pixel value of 255 and the other pixel values of 0. In addition, by using appropriate parameters of statistical region merging algorithm, segmentation comparison is performed. In the step of feature extraction, the feature vectors belonging to the vehicles are obtained by using Gabor filters with different orientation, frequency and angle values. A number of Gabor filters are created by changing the orientation, frequency and angle parameters of the Gabor filters to extract important features of the images that form the distinctive parts. Finally, images are classified by sparse representation method. In the study, l₁ norm analysis of sparse representation is used. A joint database of the feature vectors generated by the target images of military vehicle types is obtained side by side and this database is transformed into the matrix form. In order to classify the vehicles in a similar way, the test images of each vehicle is converted to the vector form and l₁ norm analysis of the sparse representation method is applied through the existing database matrix form. As a result, correct recognition has been performed by matching the target images of military vehicles with the test images by means of the sparse representation method. 97% classification success of SAR images of different military vehicle types is obtained.

Keywords: automatic target recognition, sparse representation, image classification, SAR images

Procedia PDF Downloads 365

Authors: Shih-Wen Hsiao, Rong-Qi Chen, Chien-Yu Lin

Abstract:

The topic of surface flattening plays a vital role in the field of computer aided design and manufacture. Surface flattening enables the production of 2D patterns and it can be used in design and manufacturing for developing a 3D surface to a 2D platform, especially in fashion design. This study describes surface flattening based on minimum energy methods according to the property of different fabrics. Firstly, through the geometric feature of a 3D surface, the less transformed area can be flattened on a 2D platform by geodesic. Then, strain energy that has accumulated in mesh can be stably released by an approximate implicit method and revised error function. In some cases, cutting mesh to further release the energy is a common way to fix the situation and enhance the accuracy of the surface flattening, and this makes the obtained 2D pattern naturally generate significant cracks. When this methodology is applied to a 3D mannequin constructed with feature lines, it enhances the level of computer-aided fashion design. Besides, when different fabrics are applied to fashion design, it is necessary to revise the shape of a 2D pattern according to the properties of the fabric. With this model, the outline of 2D patterns can be revised by distributing the strain energy with different results according to different fabric properties. Finally, this research uses some common design cases to illustrate and verify the feasibility of this methodology.

Keywords: surface flattening, strain energy, minimum energy, approximate implicit method, fashion design

Procedia PDF Downloads 334

Authors: Hoda Ghavamipoor, S. Alireza Hashemi Golpayegani

Abstract:

An approach to model the customer interaction with e-commerce websites is presented. Considering the service quality level as a predictive feature, we offer an improved method based on the Customer Behavior Model Graph (CBMG), a state-transition graph model. To derive the Quality of Service sensitive-CBMG (QoS-CBMG) model, process-mining techniques is applied to pre-processed website server logs which are categorized as ‘buy’ or ‘visit’. Experimental results on an e-commerce website data confirmed that the proposed method outperforms CBMG based method.

Keywords: customer behavior model, electronic commerce, quality of service, customer behavior model graph, process mining

Procedia PDF Downloads 416

Authors: Bhalke D. G., Bormane D. S., Kharate G. K.

Abstract:

This paper presents classification of musical instrument using machine learning techniques. The classification has been carried out using temporal, spectral, cepstral and wavelet features. Detail feature analysis is carried out using separate and combined features. Further, instrument model has been developed using K-Nearest Neighbor and Support Vector Machine (SVM). Benchmarked McGill university database has been used to test the performance of the system. Experimental result shows that SVM performs better as compared to KNN classifier.

Keywords: feature extraction, SVM, KNN, musical instruments

Procedia PDF Downloads 480

Authors: Imed Feki, Faouzi Msahli

Abstract:

Selection of relevant parameters from a high dimensional process operation setting space is a problem frequently encountered in industrial process modelling. This paper presents a method for selecting the most relevant fabric physical parameters for each sensory quality feature. The proposed relevancy criterion has been developed using two approaches. The first utilizes a fuzzy sensitivity criterion by exploiting from experimental data the relationship between physical parameters and all the sensory quality features for each evaluator. Next an OWA aggregation procedure is applied to aggregate the ranking lists provided by different evaluators. In the second approach, another panel of experts provides their ranking lists of physical features according to their professional knowledge. Also by applying OWA and a fuzzy aggregation model, the data sensitivity-based ranking list and the knowledge-based ranking list are combined using our proposed percolation technique, to determine the final ranking list. The key issue of the proposed percolation technique is to filter automatically and objectively the relevant features by creating a gap between scores of relevant and irrelevant parameters. It permits to automatically generate threshold that can effectively reduce human subjectivity and arbitrariness when manually choosing thresholds. For a specific sensory descriptor, the threshold is defined systematically by iteratively aggregating (n times) the ranking lists generated by OWA and fuzzy models, according to a specific algorithm. Having applied the percolation technique on a real example, of a well known finished textile product especially the stonewashed denims, usually considered as the most important quality criteria in jeans’ evaluation, we separate the relevant physical features from irrelevant ones for each sensory descriptor. The originality and performance of the proposed relevant feature selection method can be shown by the variability in the number of physical features in the set of selected relevant parameters. Instead of selecting identical numbers of features with a predefined threshold, the proposed method can be adapted to the specific natures of the complex relations between sensory descriptors and physical features, in order to propose lists of relevant features of different sizes for different descriptors. In order to obtain more reliable results for selection of relevant physical features, the percolation technique has been applied for combining the fuzzy global relevancy and OWA global relevancy criteria in order to clearly distinguish scores of the relevant physical features from those of irrelevant ones.

Keywords: data sensitivity, feature selection, fuzzy logic, OWA operators, percolation technique

Procedia PDF Downloads 605

Authors: Gaetano Zazzaro, Angelo Martone, Roberto V. Montaquila, Luigi Pavone

Abstract:

Seizure is the main factor that affects the quality of life of epileptic patients. The diagnosis of epilepsy, and hence the identification of epileptogenic zone, is commonly made by using continuous Electroencephalogram (EEG) signal monitoring. Seizure identification on EEG signals is made manually by epileptologists and this process is usually very long and error prone. The aim of this paper is to describe an automated method able to detect seizures in EEG signals, using knowledge discovery in database process and data mining methods and algorithms, which can support physicians during the seizure detection process. Our detection method is based on Artificial Neural Network classifier, trained by applying the multilayer perceptron algorithm, and by using a software application, called Training Builder that has been developed for the massive extraction of features from EEG signals. This tool is able to cover all the data preparation steps ranging from signal processing to data analysis techniques, including the sliding window paradigm, the dimensionality reduction algorithms, information theory, and feature selection measures. The final model shows excellent performances, reaching an accuracy of over 99% during tests on data of a single patient retrieved from a publicly available EEG dataset.

Keywords: artificial neural network, data mining, electroencephalogram, epilepsy, feature extraction, seizure detection, signal processing

Procedia PDF Downloads 188

Authors: Jianhong Xiang, Rui Sun, Linyu Wang

Abstract:

In recent years, due to the continuous improvement of deep learning theory, Convolutional Neural Network (CNN) has played a great superior performance in the research of Hyperspectral Image (HSI) classification. Since HSI has rich spatial-spectral information, only utilizing a single dimensional or single size convolutional kernel will limit the detailed feature information received by CNN, which limits the classification accuracy of HSI. In this paper, we design a multi-scale CNN with MLP based on residual and attention modules (INRAM-3DCNN) for the HSI classification task. We propose to use multiple 3D convolutional kernels to extract the packet feature information and fully learn the spatial-spectral features of HSI while designing residual 3D convolutional branches to avoid the decline of classification accuracy due to network degradation. Secondly, we also design the 2D Inception module with a joint channel attention mechanism to quickly extract key spatial feature information at different scales of HSI and reduce the complexity of the 3D model. Due to the high parallel processing capability and nonlinear global action of the Multilayer Perceptron (MLP), we use it in combination with the previous CNN structure for the final classification process. The experimental results on two HSI datasets show that the proposed INRAM-3DCNN method has superior classification performance and can perform the classification task excellently.

Keywords: INRAM-3DCNN, residual, channel attention, hyperspectral image classification

Procedia PDF Downloads 79

Authors: Norah Mohammed Alshahrani, Abdulaziz Almaleh

Abstract:

Autism Spectrum Disorder (ASD) is a condition related to issues with brain development that affects how a person recognises and communicates with others which results in difficulties with interaction and communication socially and it is constantly growing. Early recognition of ASD allows children to lead safe and healthy lives and helps doctors with accurate diagnoses and management of conditions. Therefore, it is crucial to develop a method that will achieve good results and with high accuracy for the measurement of ASD in children. In this paper, ASD datasets of toddlers and children have been analyzed. We employed the following machine learning techniques to attempt to explore ASD and they are Random Forest (RF), Decision Tree (DT), Na¨ıve Bayes (NB) and Support Vector Machine (SVM). Then Feature selection was used to provide fewer attributes from ASD datasets while preserving model performance. As a result, we found that the best result has been provided by the Support Vector Machine (SVM), achieving 0.98% in the toddler dataset and 0.99% in the children dataset.

Keywords: autism spectrum disorder, machine learning, feature selection, support vector machine

Procedia PDF Downloads 151

Authors: Y. Xu, L. Xiong, Z. Xu

Abstract:

In order to protect data privacy, image with sensitive or private information needs to be encrypted before being outsourced to the cloud. However, this causes difficulties in image retrieval and data management. A secure image retrieval method based on orthogonal decomposition is proposed in the paper. The image is divided into two different components, for which encryption and feature extraction are executed separately. As a result, cloud server can extract features from an encrypted image directly and compare them with the features of the queried images, so that the user can thus obtain the image. Different from other methods, the proposed method has no special requirements to encryption algorithms. Experimental results prove that the proposed method can achieve better security and better retrieval precision.

Keywords: secure image retrieval, secure search, orthogonal decomposition, secure cloud computing

Procedia PDF Downloads 484

Authors: A. N. Paithane, D. S. Bormane, S. D. Shirbahadurkar

Abstract:

It has been seen that emotion recognition is an important research topic in the field of Human and computer interface. A novel technique for Feature Extraction (FE) has been presented here, further a new method has been used for human emotion recognition which is based on HHT method. This method is feasible for analyzing the nonlinear and non-stationary signals. Each signal has been decomposed into the IMF using the EMD. These functions are used to extract the features using fission and fusion process. The decomposition technique which we adopt is a new technique for adaptively decomposing signals. In this perspective, we have reported here potential usefulness of EMD based techniques.We evaluated the algorithm on Augsburg University Database; the manually annotated database.

Keywords: intrinsic mode function (IMF), Hilbert-Huang transform (HHT), empirical mode decomposition (EMD), emotion detection, electrocardiogram (ECG)

Procedia PDF Downloads 580

Authors: Shih-Wen Hsiao, Chu-Hsuan Lee, Rong-Qi Chen

Abstract:

When it comes to last, it is regarded as the critical foundation of shoe design and development. Not only the last relates to the comfort of shoes wearing but also it aids the production of shoe styling and manufacturing. In order to enhance the efficiency and application of last development, a computer aided methodology for customized last form designs is proposed in this study. The reverse engineering is mainly applied to the process of scanning for the last form. Then the minimum energy is used for the revision of surface continuity, the surface of the last is reconstructed with the feature curves of the scanned last. When the surface of a last is reconstructed, based on the foundation of the proposed last form reconstruction module, the weighted arithmetic mean method is applied to the calculation on the shape morphing which differs from the grading for the control mesh of last, and the algorithm of subdivision is used to create the surface of last mesh, thus the feet-fitting 3D last form of different sizes is generated from its original form feature with functions remained. Finally, the practicability of the proposed methodology is verified through later case studies.

Keywords: 3D last design, customization, reverse engineering, weighted morphing, shape blending

Procedia PDF Downloads 339

Authors: Shih-Wen Hsiao, Rong-Qi Chen

Abstract:

In the field of fashion design, 3D Mannequin is a kind of assisting tool which could rapidly realize the design concepts. While the concept of 3D Mannequin is applied to the computer added fashion design, it will connect with the development and the application of design platform and system. Thus, the situation mentioned above revealed a truth that it is very critical to develop a module of 3D Mannequin which would correspond with the necessity of fashion design. This research proposes a concrete plan that developing and constructing a system of 3D Mannequin with Kinect. In the content, ergonomic measurements of objective human features could be attained real-time through the implement with depth camera of Kinect, and then the mesh morphing can be implemented through transformed the locations of the control-points on the model by inputting those ergonomic data to get an exclusive 3D mannequin model. In the proposed methodology, after the scanned points from the Kinect are revised for accuracy and smoothening, a complete human feature would be reconstructed by the ICP algorithm with the method of image processing. Also, the objective human feature could be recognized to analyze and get real measurements. Furthermore, the data of ergonomic measurements could be applied to shape morphing for the division of 3D Mannequin reconstructed by feature curves. Due to a standardized and customer-oriented 3D Mannequin would be generated by the implement of subdivision, the research could be applied to the fashion design or the presentation and display of 3D virtual clothes. In order to examine the practicality of research structure, a system of 3D Mannequin would be constructed with JAVA program in this study. Through the revision of experiments the practicability-contained research result would come out.

Keywords: 3D mannequin, kinect scanner, interactive closest point, shape morphing, subdivision

Procedia PDF Downloads 306

Authors: Vincent Liu

Abstract:

Machine learning algorithms (ML) can achieve high accuracy in predicting outcomes compared to classical models. Metaheuristic, nature-inspired algorithms can enhance traditional ML algorithms by optimizing them such as by performing feature selection. We compare ten ML algorithms to predict 30-day hospital readmission rates for diabetes patients in the US using a dataset from UCI Machine Learning Repository with feature selection performed by Greywolf nature-inspired algorithm. The baseline accuracy for the initial random forest model was 65%. After performing feature engineering, SMOTE for class balancing, and Greywolf optimization, the machine learning algorithms showed better metrics, including F1 scores, accuracy, and confusion matrix with improvements ranging in 10%-30%, and a best model of XGBoost with an accuracy of 95%. Applying machine learning this way can improve patient outcomes as unnecessary rehospitalizations can be prevented by focusing on patients that are at a higher risk of readmission.

Keywords: diabetes, machine learning, 30-day readmission, metaheuristic

Procedia PDF Downloads 61

Authors: Getaneh Berie Tarekegn

Abstract:

A precise localization system is crucial for many artificial intelligence Internet of Things (AI-IoT) applications in the era of smart cities. Their applications include traffic monitoring, emergency alarming, environmental monitoring, location-based advertising, intelligent transportation, and smart health care. The most common method for providing continuous positioning services in outdoor environments is by using a global navigation satellite system (GNSS). Due to nonline-of-sight, multipath, and weather conditions, GNSS systems do not perform well in dense urban, urban, and suburban areas.This paper proposes a generative AI-based positioning scheme for large-scale wireless settings using fingerprinting techniques. In this article, we presented a novel semi-supervised deep convolutional generative adversarial network (S-DCGAN)-based radio map construction method for real-time device localization. We also employed a reliable signal fingerprint feature extraction method with t-distributed stochastic neighbor embedding (t-SNE), which extracts dominant features while eliminating noise from hybrid WLAN and long-term evolution (LTE) fingerprints. The proposed scheme reduced the workload of site surveying required to build the fingerprint database by up to 78.5% and significantly improved positioning accuracy. The results show that the average positioning error of GAILoc is less than 39 cm, and more than 90% of the errors are less than 82 cm. That is, numerical results proved that, in comparison to traditional methods, the proposed SRCLoc method can significantly improve positioning performance and reduce radio map construction costs.

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

Procedia PDF Downloads 71

Authors: Yanting Cao, Kazumitsu Nawata

Abstract:

As smartphones upgrading, their software and hardware are getting smarter, so the smartphone-based human activity recognition will be described as more refined, complex, and detailed. In this context, we analyzed a set of experimental data obtained by observing and measuring 30 volunteers with six activities of daily living (ADL). Due to the large sample size, especially a 561-feature vector with time and frequency domain variables, cleaning these intractable features and training a proper model becomes extremely challenging. After a series of feature selection and parameters adjustment, a well-performed SVM classifier has been trained.

Keywords: smart sensors, human activity recognition, artificial intelligence, SVM

Procedia PDF Downloads 144

Authors: Getaneh Berie Tarekegn

Abstract:

A precise localization system is crucial for many artificial intelligence Internet of Things (AI-IoT) applications in the era of smart cities. Their applications include traffic monitoring, emergency alarming, environmental monitoring, location-based advertising, intelligent transportation, and smart health care. The most common method for providing continuous positioning services in outdoor environments is by using a global navigation satellite system (GNSS). Due to nonline-of-sight, multipath, and weather conditions, GNSS systems do not perform well in dense urban, urban, and suburban areas.This paper proposes a generative AI-based positioning scheme for large-scale wireless settings using fingerprinting techniques. In this article, we presented a novel semi-supervised deep convolutional generative adversarial network (S-DCGAN)-based radio map construction method for real-time device localization. We also employed a reliable signal fingerprint feature extraction method with t-distributed stochastic neighbor embedding (t-SNE), which extracts dominant features while eliminating noise from hybrid WLAN and long-term evolution (LTE) fingerprints. The proposed scheme reduced the workload of site surveying required to build the fingerprint database by up to 78.5% and significantly improved positioning accuracy. The results show that the average positioning error of GAILoc is less than 39 cm, and more than 90% of the errors are less than 82 cm. That is, numerical results proved that, in comparison to traditional methods, the proposed SRCLoc method can significantly improve positioning performance and reduce radio map construction costs.

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

Procedia PDF Downloads 75

Authors: Shreya Asthana

Abstract:

Testers know that the feature they tested on stage is working perfectly in production only after release went live. Sometimes something breaks in production and testers get to know through the end user’s bug raised. The panic mode starts when your staging test results do not reflect current production behavior. And you started doubting your testing skills when finally the user reported a bug to you. Testers can deploy their confidence on release day by testing on production. Once you start doing testing in production, you will see test result accuracy because it will be running on real time data and execution will be a little faster as compared to staging one due to elimination of bad data. Feature flagging, canary releases, and data cleanup can help to achieve this technique of testing. By this paper it will be easier to understand the steps to achieve production testing before making your feature live, and to modify IT company’s testing procedure, so testers can provide the bug free experience to the end users. This study is beneficial because too many people think that testing should be done in staging but not in production and now this is high time to pull out people from their old mindset of testing into a new testing world. At the end of the day, it all just matters if the features are working in production or not.

Keywords: bug free production, new testing mindset, testing strategy, testing approach

Procedia PDF Downloads 77

Authors: Chengming Li, Yong Yin, Peipei Guo, Xiaoli Liu

Abstract:

Without taking account of the attribute richness of POI (point of interest) data and spatial distribution limited by roads, a POI generalization method considering both attribute information and spatial distribution has been proposed against the existing point generalization algorithm merely focusing on overall information of point groups. Hierarchical characteristic of urban POI information expression has been firstly analyzed to point out the measurement feature of the corresponding hierarchy. On this basis, an urban POI generalizing strategy has been put forward: POIs urban road network have been divided into three distribution pattern; corresponding generalization methods have been proposed according to the characteristic of POI data in different distribution patterns. Experimental results showed that the method taking into account both attribute information and spatial distribution characteristics of POI can better implement urban POI generalization in the mapping presentation.

Keywords: POI, road network, selection method, spatial information expression, distribution pattern

Procedia PDF Downloads 410

Authors: A. Ashok, K.Satapathy, B. Prerana Nashine

Abstract:

The objective of this research work is to investigate for one dimensional transient radiative transfer equations with conduction using finite volume method. Within the infrastructure of finite-volume, we obtain the conservative discretization of the terms in order to preserve the overall conservative property of finitevolume schemes. Coupling of conductive and radiative equation resulting in fluxes is governed by the magnitude of emissivity, extinction coefficient, and temperature of the medium as well as geometry of the problem. The problem under consideration has been solved, for a slab dominating radiation coupled with transient conduction based on finite volume method. The boundary conditions are also chosen so as to give a good model of the discretized form of radiation transfer equation. The important feature of the present method is flexibility in specifying the control angles in the FVM, while keeping the simplicity in the solution procedure. Effects of various model parameters are examined on the distributions of temperature, radiative and conductive heat fluxes and incident radiation energy etc. The finite volume method is considered to effectively evaluate the propagation of radiation intensity through a participating medium.

Keywords: participating media, finite volume method, radiation coupled with conduction, transient radiative heat transfer

Procedia PDF Downloads 389

Authors: Xun Li, Haojie Wang

Abstract:

Federated learning is a kind of distributed training and centralized training mode, which is of great value in the protection of user privacy. In order to solve the problem that the model is vulnerable to backdoor attacks in federated learning, a backdoor attack detection method based on a weighted k-medoids algorithm is proposed. First of all, this paper collates the update parameters of the client to construct a vector group, then uses the principal components analysis (PCA) algorithm to extract the corresponding feature information from the vector group, and finally uses the improved k-medoids clustering algorithm to identify the normal and backdoor update parameters. In this paper, the backdoor is implanted in the federation learning model through the model replacement attack method in the simulation experiment, and the update parameters from the attacker are effectively detected and removed by the defense method proposed in this paper.

Keywords: federated learning, backdoor attack, PCA, k-medoids, backdoor defense

Procedia PDF Downloads 114

Authors: Chin-Chen Chang, I-Ta Lee, Tsung-Ta Ke, Wen-Kai Tai

Abstract:

Common methods for resizing image size include scaling and cropping. However, these two approaches have some quality problems for reduced images. In this paper, we propose an image resizing algorithm by separating the main objects and the background. First, we extract two feature maps, namely, an enhanced visual saliency map and an improved gradient map from an input image. After that, we integrate these two feature maps to an importance map. Finally, we generate the target image using the importance map. The proposed approach can obtain desired results for a wide range of images.

Keywords: energy map, visual saliency, gradient map, seam carving

Procedia PDF Downloads 476

Authors: Sung Un Kim, Ivana Martinović, Snježana Stanarević Katavić

Abstract:

As more and more youth use mobile devices, such as tablets and smartphones, for information seeking in their everyday lives, the purpose of this study is to understand the behaviors of youth seeking health information on mobile devices. The specific objective of this study is to examine 1) for what health issues youth use mobile devices, 2) for what reasons youth use mobile devices to obtain health information, 3) in what ways youth use mobile devices for health information, and 4) the features of health applications that youth find useful. The researchers devised a questionnaire for this study. Four hundred eight students from two high schools, located in Osijek, Croatia, participated by answering the questionnaire (281 girls and 127 boys). The collected data were analyzed using descriptive statistics and content analysis. The results show that among all participants, about 85 percent (n = 344) reported having used mobile devices for health information. The most frequent health topic for which they had been using mobile devices is physical activity (n = 273), followed by eating issues and nutrition (n = 224), mental health (n = 160), sexual health (n = 157), alcohol, drugs, and tobacco (n = 125), safety (n = 96) and particular diseases (n = 62). They use mobile devices to obtain health information due to the ease of use (n = 342), the ease of sharing health information (n = 281), portability (n = 215), timeliness (n = 162), and the ease of tracking/recording/monitoring health status (n = 147). Of those who have used mobile devices for health information, three-quarters (n = 261) use mobile devices to search health information, while 32.8% (n =113) use applications and 31.7% (n =109) browse information. Those who have used applications for health information (n = 113) consider the alert feature (n=107) as the most useful, followed by the tracking/recording/monitoring feature (n =92), the customized information feature (n = 86), the video feature (n = 58), and the sharing feature (n =39). It is notable that although health applications have been actively developed and studied, a majority of the participants search for or browse information on mobile devices, instead of using applications. The researchers will discuss reasons that some of them did not use mobile devices to obtain health information, students’ concerns about using health applications, and features that they wish to have in health applications.

Keywords: Croatia, health information, information seeking behaviors, mobile devices, youth

Procedia PDF Downloads 402

Authors: Shibo Wei, Ting Jiang

Abstract:

Convolutional-recurrent neural networks (CRN) have achieved much success recently in the speech enhancement field. The common processing method is to use the convolution layer to compress the feature space by multiple upsampling and then model the compressed features with the LSTM layer. At last, the enhanced speech is obtained by deconvolution operation to integrate the global information of the speech sequence. However, the feature space compression process may cause the loss of information, so we propose to model the upsampling result of each step with the residual LSTM layer, then join it with the output of the deconvolution layer and input them to the next deconvolution layer, by this way, we want to integrate the global information of speech sequence better. The experimental results show the network model (RES-CRN) we introduce can achieve better performance than LSTM without residual and overlaying LSTM simply in the original CRN in terms of scale-invariant signal-to-distortion ratio (SI-SNR), speech quality (PESQ), and intelligibility (STOI).

Keywords: convolutional-recurrent neural networks, speech enhancement, residual LSTM, SI-SNR

Procedia PDF Downloads 200

Authors: Umm-E-Hani, Faiza Babar, Hanif Durad

Abstract:

Malware-detecting approaches suffer many limitations, due to which all anti-malware solutions have failed to be reliable enough for detecting zero-day malware. Signature-based solutions depend upon the signatures that can be generated only when malware surfaces at least once in the cyber world. Another approach that works by detecting the anomalies caused in the environment can easily be defeated by diligently and intelligently written malware. Solutions that have been trained to observe the behavior for detecting malicious files have failed to cater to the malware capable of detecting the sandboxed or protected environment. Machine learning and deep learning-based approaches greatly suffer in training their models with either an imbalanced dataset or an inadequate number of samples. AI-based anti-malware solutions that have been trained with enough samples targeted a selected feature vector, thus ignoring the input of leftover features in the maliciousness of malware just to cope with the lack of underlying hardware processing power. Our research focuses on producing an anti-malware solution for detecting malicious PE files by circumventing the earlier-mentioned shortcomings. Our proposed framework, which is based on automated feature engineering through autoencoders, trains the model over a fairly large dataset. It focuses on the visual patterns of malware samples to automatically extract the meaningful part of the visual pattern. Our experiment has successfully produced a state-of-the-art accuracy of 99.54 % over test data.

Keywords: malware, auto encoders, automated feature engineering, classification

Procedia PDF Downloads 72

Authors: Rahma Abed, Sahbi Bahroun, Ezzeddine Zagrouba

Abstract:

Due to the huge amount of data in videos, extracting the relevant frames became a necessity and an essential step prior to performing face recognition. In this context, we propose a method for extracting keyframes from videos based on face quality and deep learning for a face recognition task. This method has two steps. We start by generating face quality scores for each face image based on the use of three face feature extractors, including Gabor, LBP, and HOG. The second step consists in training a Deep Convolutional Neural Network in a supervised manner in order to select the frames that have the best face quality. The obtained results show the effectiveness of the proposed method compared to the methods of the state of the art.

Keywords: keyframe extraction, face quality assessment, face in video recognition, convolution neural network

Procedia PDF Downloads 233

Authors: Thi Thi Zin, Pyke Tin, Ikuo Kobayashi, Yoichiro Horii

Abstract:

Today dairy farm experts and farmers have well recognized the importance of dairy cow Body Condition Score (BCS) since these scores can be used to optimize milk production, managing feeding system and as an indicator for abnormality in health even can be utilized to manage for having healthy calving times and process. In tradition, BCS measures are done by animal experts or trained technicians based on visual observations focusing on pin bones, pin, thurl and hook area, tail heads shapes, hook angles and short and long ribs. Since the traditional technique is very manual and subjective, the results can lead to different scores as well as not cost effective. Thus this paper proposes an algebraic geometric imaging approach for an automatic dairy cow BCS system. The proposed system consists of three functional modules. In the first module, significant landmarks or anatomical points from the cow image region are automatically extracted by using image processing techniques. To be specific, there are 23 anatomical points in the regions of ribs, hook bones, pin bone, thurl and tail head. These points are extracted by using block region based vertical and horizontal histogram methods. According to animal experts, the body condition scores depend mainly on the shape structure these regions. Therefore the second module will investigate some algebraic and geometric properties of the extracted anatomical points. Specifically, the second order polynomial regression is employed to a subset of anatomical points to produce the regression coefficients which are to be utilized as a part of feature vector in scoring process. In addition, the angles at thurl, pin, tail head and hook bone area are computed to extend the feature vector. Finally, in the third module, the extracted feature vectors are trained by using Markov Classification process to assign BCS for individual cows. Then the assigned BCS are revised by using multiple regression method to produce the final BCS score for dairy cows. In order to confirm the validity of proposed method, a monitoring video camera is set up at the milk rotary parlor to take top view images of cows. The proposed method extracts the key anatomical points and the corresponding feature vectors for each individual cows. Then the multiple regression calculator and Markov Chain Classification process are utilized to produce the estimated body condition score for each cow. The experimental results tested on 100 dairy cows from self-collected dataset and public bench mark dataset show very promising with accuracy of 98%.

Keywords: algebraic geometric imaging approach, body condition score, Markov classification, polynomial regression

Procedia PDF Downloads 157

Authors: Ali Cheraghian, Farshid Hajati, Soheila Gheisari, Yongsheng Gao

Abstract:

In this paper, we present a novel 2.5D face recognition method based on Gabor Discrete Cosine Transform (GDCT). In the proposed method, the Gabor filter is applied to extract feature vectors from the texture and the depth information. Then, Discrete Cosine Transform (DCT) is used for dimensionality and redundancy reduction to improve computational efficiency. The system is combined texture and depth information in the decision level, which presents higher performance compared to methods, which use texture and depth information, separately. The proposed algorithm is examined on publically available Bosphorus database including models with pose variation. The experimental results show that the proposed method has a higher performance compared to the benchmark.

Keywords: Gabor filter, discrete cosine transform, 2.5d face recognition, pose

Procedia PDF Downloads 328

Authors: Isil Kerti, Onur Okur, Sibel Daglilar, Recep Calin

Abstract:

Particulate reinforced aluminium matrix composites have gained more importance in automotive, aeronautical and defense industries due to their specific properties like as low density, high strength and stiffness, good fatigue strength, dimensional stability at high temperature and acceptable tribological properties. In this study, 2014 Aluminium alloy used as a matrix material and B₄C and SiC were selected as reinforcements components. For production of composites materials, vacuum infiltration method was used. In the experimental studies, the reinforcement volume ratios were defined by mixing as totally 10% B₄C and SiC. Aging treatment (T6) was applied to the specimens. The effect of T6 treatment on hardness was determined by using Brinell hardness test method. The effects of the aging treatment on microstructure and chemical structure were analysed by making XRD, SEM and EDS analysis on the specimens.

Keywords: metal matrix composite, vacumm infiltration method, aluminum metal matrix, mechanical feature

Procedia PDF Downloads 315

Authors: Hayriye Anıl, Görkem Kar

Abstract:

In addition to being rich in renewable energy resources, Turkey is one of the countries that promise potential in geothermal energy production with its high installed power, cheapness, and sustainability. Increasing imbalance penalties become an economic burden for organizations since geothermal generation plants cannot maintain the balance of supply and demand due to the inadequacy of the production forecasts given in the day-ahead market. A better production forecast reduces the imbalance penalties of market participants and provides a better imbalance in the day ahead market. In this study, using machine learning, deep learning, and, time series methods, the total generation of the power plants belonging to Zorlu Natural Electricity Generation, which has a high installed capacity in terms of geothermal, was estimated for the first one and two weeks of March, then the imbalance penalties were calculated with these estimates and compared with the real values. These modeling operations were carried out on two datasets, the basic dataset and the dataset created by extracting new features from this dataset with the feature engineering method. According to the results, Support Vector Regression from traditional machine learning models outperformed other models and exhibited the best performance. In addition, the estimation results in the feature engineering dataset showed lower error rates than the basic dataset. It has been concluded that the estimated imbalance penalty calculated for the selected organization is lower than the actual imbalance penalty, optimum and profitable accounts.

Keywords: machine learning, deep learning, time series models, feature engineering, geothermal energy production forecasting

Procedia PDF Downloads 110