Search results for: graph recognition

Authors: Taysir Soliman

Abstract:

One of the current popular clustering techniques is Spectral Clustering (SC) because of its advantages over conventional approaches such as hierarchical clustering, k-means, etc. and other techniques as well. However, one of the disadvantages of SC is the time consuming process because it requires computing the eigenvectors. In the past to overcome this disadvantage, a number of attempts have been proposed such as the Power Iteration Clustering (PIC) technique, which is one of versions from SC; some of PIC advantages are: 1) its scalability and efficiency, 2) finding one pseudo-eigenvectors instead of computing eigenvectors, and 3) linear combination of the eigenvectors in linear time. However, its worst disadvantage is an inter-class collision problem because it used only one pseudo-eigenvectors which is not enough. Previous researchers developed Deflation-based Power Iteration Clustering (DPIC) to overcome problems of PIC technique on inter-class collision with the same efficiency of PIC. In this paper, we developed Parallel DPIC (PDPIC) to improve the time and memory complexity which is run on apache spark framework using sparse matrix. To test the performance of PDPIC, we compared it to SC, ESCG, ESCALG algorithms on four small graph benchmark datasets and nine large graph benchmark datasets, where PDPIC proved higher accuracy and better time consuming than other compared algorithms.

Keywords: spectral clustering, power iteration clustering, deflation-based power iteration clustering, Apache spark, large graph

Procedia PDF Downloads 189

Authors: Keyi Wang

Abstract:

Similar to the touchscreen, hand gesture based human-computer interaction (HCI) is a technology that could allow people to perform a variety of tasks faster and more conveniently. This paper proposes a training method of an image-based hand gesture image and video clip recognition system using a CNN (Convolutional Neural Network) with a dataset. A dataset containing 6 hand gesture images is used to train a 2D CNN model. ~98% accuracy is achieved. Furthermore, a 3D CNN model is trained on a dataset containing 4 hand gesture video clips resulting in ~83% accuracy. It is demonstrated that a Cozmo robot loaded with pre-trained models is able to recognize static and dynamic hand gestures.

Keywords: deep learning, hand gesture recognition, computer vision, image processing

Procedia PDF Downloads 139

Authors: P. Shrivastava, A. Shukla, K. Verma, S. Rungta

Abstract:

Parkinson's disease is a chronic neurological disorder that directly affects human gait. It leads to slowness of movement, causes muscle rigidity and tremors. Gait serve as a primary outcome measure for studies aiming at early recognition of disease. Using gait techniques, this paper implements efficient binary bat algorithm for an early detection of Parkinson's disease by selecting optimal features required for classification of affected patients from others. The data of 166 people, both fit and affected is collected and optimal feature selection is done using PSO and Bat algorithm. The reduced dataset is then classified using neural network. The experiments indicate that binary bat algorithm outperforms traditional PSO and genetic algorithm and gives a fairly good recognition rate even with the reduced dataset.

Keywords: parkinson, gait, feature selection, bat algorithm

Procedia PDF Downloads 545

Authors: Dawood Al Chanti, Alice Caplier

Abstract:

Sparse representations of signals have received a great deal of attention in recent years. In this paper, the interest of using sparse representation as a mean for performing sparse discriminative analysis between spontaneous facial expressions is demonstrated. An automatic facial expressions recognition system is presented. It uses a KSVD-SVM approach which is made of three main stages: A pre-processing and feature extraction stage, which solves the problem of shared subspace distribution based on the random projection theory, to obtain low dimensional discriminative and reconstructive features; A dictionary learning and sparse coding stage, which uses the KSVD model to learn discriminative under or over dictionaries for sparse coding; Finally a classification stage, which uses a SVM classifier for facial expressions recognition. Our main concern is to be able to recognize non-basic affective states and non-acted expressions. Extensive experiments on the JAFFE static acted facial expressions database but also on the DynEmo dynamic spontaneous facial expressions database exhibit very good recognition rates.

Keywords: dictionary learning, random projection, pose and spontaneous facial expression, sparse representation

Procedia PDF Downloads 305

Authors: Omar Dib, Alexandre Caminada, Marie-Ange Manier

Abstract:

The purpose of this study is to introduce a novel approach to solve the one-to-one shortest path problem. A directed connected graph is assumed in which all edges’ weights are positive. Our method is based on a memetic algorithm in which we combine a genetic algorithm (GA) and a variable neighborhood search method (VNS). We compare our approximate method with two exact algorithms Dijkstra and Integer Programming (IP). We made experimentations using random generated, complete and real graph instances. In most case studies, numerical results show that our method outperforms exact methods with 5% average gap to the optimality. Our algorithm’s average speed is 20-times faster than Dijkstra and more than 1000-times compared to IP. The details of the experimental results are also discussed and presented in the paper.

Keywords: shortest path problem, Dijkstra’s algorithm, integer programming, memetic algorithm

Procedia PDF Downloads 466

Authors: Rabab M. Ramadan, Elaraby A. Elgallad

Abstract:

With extensive application, the performance of unimodal biometrics systems has to face a diversity of problems such as signal and background noise, distortion, and environment differences. Therefore, multimodal biometric systems are proposed to solve the above stated problems. This paper introduces a bimodal biometric recognition system based on the extracted features of the human palm print and iris. Palm print biometric is fairly a new evolving technology that is used to identify people by their palm features. The iris is a strong competitor together with face and fingerprints for presence in multimodal recognition systems. In this research, we introduced an algorithm to the combination of the palm and iris-extracted features using a texture-based descriptor, the Scale Invariant Feature Transform (SIFT). Since the feature sets are non-homogeneous as features of different biometric modalities are used, these features will be concatenated to form a single feature vector. Particle swarm optimization (PSO) is used as a feature selection technique to reduce the dimensionality of the feature. The proposed algorithm will be applied to the Institute of Technology of Delhi (IITD) database and its performance will be compared with various iris recognition algorithms found in the literature.

Keywords: iris recognition, particle swarm optimization, feature extraction, feature selection, palm print, the Scale Invariant Feature Transform (SIFT)

Procedia PDF Downloads 235

Authors: N. N. Mosola, S. J. Molete, L. S. Masoebe, M. Letsae

Abstract:

Hand gesture recognition is a technique used to locate, detect, and recognize a hand gesture. Detection and recognition are concepts of Artificial Intelligence (AI). AI concepts are applicable in Human Computer Interaction (HCI), Expert systems (ES), etc. Hand gesture recognition can be used in sign language interpretation. Sign language is a visual communication tool. This tool is used mostly by deaf societies and those with speech disorder. Communication barriers exist when societies with speech disorder interact with others. This research aims to build a hand recognition system for Lesotho’s Sesotho and English language interpretation. The system will help to bridge the communication problems encountered by the mentioned societies. The system has various processing modules. The modules consist of a hand detection engine, image processing engine, feature extraction, and sign recognition. Detection is a process of identifying an object. The proposed system uses Canny pruning Haar and Haarcascade detection algorithms. Canny pruning implements the Canny edge detection. This is an optimal image processing algorithm. It is used to detect edges of an object. The system employs a skin detection algorithm. The skin detection performs background subtraction, computes the convex hull, and the centroid to assist in the detection process. Recognition is a process of gesture classification. Template matching classifies each hand gesture in real-time. The system was tested using various experiments. The results obtained show that time, distance, and light are factors that affect the rate of detection and ultimately recognition. Detection rate is directly proportional to the distance of the hand from the camera. Different lighting conditions were considered. The more the light intensity, the faster the detection rate. Based on the results obtained from this research, the applied methodologies are efficient and provide a plausible solution towards a light-weight, inexpensive system which can be used for sign language interpretation.

Keywords: canny pruning, hand recognition, machine learning, skin tracking

Procedia PDF Downloads 185

Authors: Abdullah A. AlShaher

Abstract:

In this paper, we demonstrate how regression curves can be used to recognize 2D non-rigid handwritten shapes. Each shape is represented by a set of non-overlapping uniformly distributed landmarks. The underlying models utilize 2^nd order of polynomials to model shapes within a training set. To estimate the regression models, we need to extract the required coefficients which describe the variations for a set of shape class. Hence, a least square method is used to estimate such modes. We then proceed by training these coefficients using the apparatus Expectation Maximization algorithm. Recognition is carried out by finding the least error landmarks displacement with respect to the model curves. Handwritten isolated Arabic characters are used to evaluate our approach.

Keywords: character recognition, regression curves, handwritten Arabic letters, expectation maximization algorithm

Procedia PDF Downloads 145

Authors: Thuy Bui Anh, Ngan Nguyen Thi Thanh

Abstract:

Currently, social work in Vietnam is entering the first step in the development process to become a true profession with a strong position in society. However, Spirit of helping and sharing of social work has already existed in the daily life of Vietnamese people for a very long time, becoming a precious heritage passed down from ancestors to the next generations while expanding the territory, building and defending for the country. Following the stream of history, charity work in Vietnam has gradually transformed itself towards a more professional work, especially in the last 2 decades. Accordingly, more than 50 universities and educational institutions in Vietnam have been licensed to train social work, ensuring a stronger foundation on human resources working in this field. Despite the strong growth, social work profession, social work education and the recognition of the role of the social workers still need to be fueled to develop, responded to the increasing demand of Vietnam society.

Keywords: education, history, recognition, social work, Vietnam

Procedia PDF Downloads 319

Authors: Muhammad Azam, Micheal Olaolu Arowolo, Fei He, Mihail Popescu, Dong Xu

Abstract:

The number of biological papers is growing quickly, which means that the number of biological pathway figures in those papers is also increasing quickly. Each pathway figure shows extensive biological information, like the names of genes and how the genes are related. However, manually annotating pathway figures takes a lot of time and work. Even though using advanced image understanding models could speed up the process of curation, these models still need to be made more accurate. To improve gene name recognition from pathway figures, we applied a Siamese network to map image segments to a library of pictures containing known genes in a similar way to person recognition from photos in many photo applications. We used a triple loss function and a triplet spatial pyramid pooling network by combining the triplet convolution neural network and the spatial pyramid pooling (TSPP-Net). We compared VGG19 and VGG16 as the Siamese network model. VGG16 achieved better performance with an accuracy of 93%, which is much higher than OCR results.

Keywords: biological pathway, image understanding, gene name recognition, object detection, Siamese network, VGG

Procedia PDF Downloads 291

Authors: Henni Sid Ahmed, Belbachir Mohamed Faouzi, Jean Caelen

Abstract:

Behavior recognition has been studied for realizing drivers assisting system and automated navigation and is an important studied field in the intelligent Building. In this paper, a recognition method of behavior recognition separated from a real image was studied. Images were divided into several categories according to the actual weather, distance and angle of view etc. SIFT was firstly used to detect key points and describe them because the SIFT (Scale Invariant Feature Transform) features were invariant to image scale and rotation and were robust to changes in the viewpoint and illumination. My goal is to develop a robust and reliable system which is composed of two fixed cameras in every room of intelligent building which are connected to a computer for acquisition of video sequences, with a program using these video sequences as inputs, we use SIFT represented different images of video sequences, and SVM (support vector machine) Lights as a programming tool for classification of images in order to classify people’s behaviors in the intelligent building in order to give maximum comfort with optimized energy consumption.

Keywords: video analysis, people behavior, intelligent building, classification

Procedia PDF Downloads 378

Authors: Huy Hieu Pham, Houssam Salmane, Louahdi Khoudour, Alain Crouzil, Pablo Zegers, Sergio Velastin

Abstract:

We present a deep learning-based multitask framework for joint 3D human pose estimation and action recognition from color video sequences. Our approach proceeds along two stages. In the first, we run a real-time 2D pose detector to determine the precise pixel location of important key points of the body. A two-stream neural network is then designed and trained to map detected 2D keypoints into 3D poses. In the second, we deploy the Efficient Neural Architecture Search (ENAS) algorithm to find an optimal network architecture that is used for modeling the Spatio-temporal evolution of the estimated 3D poses via an image-based intermediate representation and performing action recognition. Experiments on Human3.6M, Microsoft Research Redmond (MSR) Action3D, and Stony Brook University (SBU) Kinect Interaction datasets verify the effectiveness of the proposed method on the targeted tasks. Moreover, we show that our method requires a low computational budget for training and inference.

Keywords: human action recognition, pose estimation, D-CNN, deep learning

Procedia PDF Downloads 146

Authors: Maha Jazouli

Abstract:

Suicide is one of the most important causes of death in the prison environment, both in Canada and internationally. Rates of attempts of suicide and self-harm have been on the rise in recent years, with hangings being the most frequent method resorted to. The objective of this article is to propose a method to automatically detect in real time suicidal behaviors. We present a gesture recognition system that consists of three modules: model-based movement tracking, feature extraction, and gesture recognition using machine learning algorithms (MLA). Our proposed system gives us satisfactory results. This smart video surveillance system can help assist staff responsible for the safety and health of inmates by alerting them when suicidal behavior is detected, which helps reduce mortality rates and save lives.

Keywords: suicide detection, Kinect azure, RGB-D camera, SVM, machine learning, gesture recognition

Procedia PDF Downloads 188

Authors: Oussama Hafferssas, Hiba Benyahia, Amina Madani, Nassima Zeriri

Abstract:

Although data nowadays has multiple forms; from text to images, and from audio to videos, yet text is still the most used one at a public level. At an academical and research level, and unlike other forms, text can be considered as the easiest form to process. Therefore, a brunch of Data Mining researches has been always under its shadow, called "Text Mining". Its concept is just like data mining’s, finding valuable patterns in data, from large collections and tremendous volumes of data, in this case: Text. Named entity recognition (NER) is one of Text Mining’s disciplines, it aims to extract and classify references such as proper names, locations, expressions of time and dates, organizations and more in a given text. Our approach "Octopub" does not aim to find new ways to improve named entity recognition process, rather than that it’s about finding a new, and yet smart way, to use NER in a way that we can extract sentiments of millions of people using Social Networks as a limitless information source, and Marketing for product promotion as the main domain of application.

Keywords: textmining, named entity recognition(NER), sentiment analysis, social media networks (SN, SMN), business intelligence(BI), marketing

Procedia PDF Downloads 589

Authors: Rafeqah Raslan, Mastura AbdManaf, Junaidah Jai, Istikamah Subuki, Ana Najwa Mustapa

Abstract:

The volatile citronella essential oil was encapsulated by simple coacervation and complex coacervation using gum Arabic and gelatin as wall material. Glutaraldehyde was used in the methodology as crosslinking agent. The citronella standard calibration graph was developed with R2 equal to 0.9523 for the accurate determination of encapsulation efficiency and release study. The release kinetic was analyzed based on Fick’s law of diffusion for polymeric system and linear graph of log fraction release over log time was constructed to determine the release rate constant, k and diffusion coefficient, n. Both coacervation methods in the present study produce encapsulation efficiency around 94%. The capsules morphology analysis supported the release kinetic mechanisms of produced capsules for both coacervation process.

Keywords: simple coacervation, complex coacervation, encapsulation efficiency, release kinetic study

Procedia PDF Downloads 316

Authors: Aderet M. Pantierer, Shmuel Pantierer, Raphael Cordina, Yougashwar Budhoo

Abstract:

Stress concentration occurs when there is an abrupt change in geometry, a mechanical part under loading. These changes in geometry can include holes, notches, or cracks within the component. The modifications create larger stress within the part. This maximum stress is difficult to determine, as it is directly at the point of the minimum area. Strain gauges have yet to be developed to analyze stresses at such minute areas. Therefore, a stress concentration factor must be utilized. The stress concentration factor is a dimensionless parameter calculated solely on the geometry of a part. The factor is multiplied by the nominal, or average, stress of the component, which can be found analytically or experimentally. Stress concentration graphs exist for common loading conditions and geometrical configurations to aid in the determination of the maximum stress a part can withstand. These graphs were developed from historical data yielded from experimentation. This project seeks to verify a stress concentration graph for combined loading conditions. The aforementioned graph was developed using CATIA Finite Element Analysis software. The results of this analysis will be validated through further testing. The 3D modeled parts will be subjected to further finite element analysis using Patran-Nastran software. The finite element models will then be verified by testing physical specimen using a tensile testing machine. Once the data is validated, the unique stress concentration graph will be submitted for publication so it can aid engineers in future projects.

Keywords: stress concentration, finite element analysis, finite element models, combined loading

Procedia PDF Downloads 443

Authors: Endrick Barnacin, Jean-Luc Henry, Jack Molinié, Jimmy Nagau, Hélène Delatte, Gérard Lebreton

Abstract:

Palynology is a field of interest for many disciplines. It has multiple applications such as chronological dating, climatology, allergy treatment, and even honey characterization. Unfortunately, the analysis of a pollen slide is a complicated and time-consuming task that requires the intervention of experts in the field, which is becoming increasingly rare due to economic and social conditions. So, the automation of this task is a necessity. Pollen slides analysis is mainly a visual process as it is carried out with the naked eye. That is the reason why a primary method to automate palynology is the use of digital image processing. This method presents the lowest cost and has relatively good accuracy in pollen retrieval. In this work, we propose a system combining recognition and grouping of pollen. It consists of using a Logistic Model Tree to classify pollen already known by the proposed system while detecting any unknown species. Then, the unknown pollen species are divided using a cluster-based approach. Success rates for the recognition of known species have been achieved, and automated clustering seems to be a promising approach.

Keywords: pollen recognition, logistic model tree, expectation-maximization, local binary pattern

Procedia PDF Downloads 182

Authors: Rutusha Bhutada, Gaurav Chavan, Sarvesh Kasat, Varsha Mujumdar

Abstract:

This software system will be an Augmented Reality application designed to maximize the doctor’s productivity by providing tools to assist in automating the patient recognition and updating patient’s records using face and voice recognition features, which would otherwise have to be performed manually. By maximizing the doctor’s work efficiency and production, the application will meet the doctor’s needs while remaining easy to understand and use. More specifically, this application is designed to allow a doctor to manage his productive time in handling the patient without losing eye-contact with him and communicate with a group of other doctors for consultation, for in-place treatments through video streaming, as a video study. The system also contains a relational database containing a list of doctor, patient and display techniques.

Keywords: augmented reality, hand-held devices, head-mounted devices, marker based systems, speech recognition, face detection

Procedia PDF Downloads 436

Authors: Jesus Alfredo Ortega Granados, Pandiyan Thangarasu

Abstract:

Antibiotic tetracycline presents as a micro-contaminant in fresh water, wastewater and soils, causing environmental and health problems. In this work, tetracycline (TC) has been employed as chemo-sensor for the recognition of Al³⁺ without interring other ions, and the results show that it enhances the fluorescence intensity for Al³⁺ and there is no interference from other coexisting cation ions (Cd²⁺, Ni²⁺, Co²⁺, Sr²⁺, Mg²⁺, Fe³⁺, K⁺, Sm³⁺, Ag⁺, Na⁺, Ba²⁺, Zn²⁺, and Mn²⁺). For the addition of Cu²⁺ to [TET-Al³⁺], it appears that the intensity of fluorescence has been quenched. Other combinations of metal ions in addition to TC do not change the fluorescence behavior. The stoichiometry determined by Job´s plot for the interaction of TC with Al³⁺ was found to be 1:1. Importantly, the detection of Al³⁺⁺ successfully employed in the real samples like living cells, and it was found that TC efficiently performs as a fluorescent probe for Al³⁺ ion in living systems, especially in Saccharomyces cerevisiae; this is confirmed by confocal laser scanning microscopy.

Keywords: chemo-sensor, recognition of Al³⁺ ion, Saccharomyces cerevisiae, tetracycline,

Procedia PDF Downloads 188

Authors: R. R. Hordijk, O. J. G. Somsen

Abstract:

Traditionally, radar systems are the eyes and ears of a ship. However, these systems have their drawbacks and nowadays they are extended with systems that work with video and photos. Processing of data from these videos and photos is however very labour-intensive and efforts are being made to automate this process. A major problem when trying to recognize objects in water is that the 'background' is not homogeneous so that traditional image recognition technics do not work well. Main question is, can a method be developed which automate this recognition process. There are a large number of parameters involved to facilitate the identification of objects on such images. One is varying the resolution. In this research, the resolution of some images has been reduced to the extreme value of 1% of the original to reduce clutter before the polynomial fit (pre-processing). It turned out that the searched object was clearly recognizable as its grey value was well above the average. Another approach is to take two images of the same scene shortly after each other and compare the result. Because the water (waves) fluctuates much faster than an object floating in the water one can expect that the object is the only stable item in the two images. Both these methods (pre-processing and comparing two images of the same scene) delivered useful results. Though it is too early to conclude that with these methods all image problems can be solved they are certainly worthwhile for further research.

Keywords: image processing, image recognition, polynomial fit, water

Procedia PDF Downloads 534

Authors: M. Jazouli, S. Elhoufi, A. Majda, A. Zarghili, R. Aalouane

Abstract:

Autism spectrum disorder is a complex developmental disability. It is defined by a certain set of behaviors. Persons with Autism Spectrum Disorders (ASD) frequently engage in stereotyped and repetitive motor movements. The objective of this article is to propose a method to automatically detect this unusual behavior. Our study provides a clinical tool which facilitates for doctors the diagnosis of ASD. We focus on automatic identification of five repetitive gestures among autistic children in real time: body rocking, hand flapping, fingers flapping, hand on the face and hands behind back. In this paper, we present a gesture recognition system for children with autism, which consists of three modules: model-based movement tracking, feature extraction, and gesture recognition using artificial neural network (ANN). The first one uses the Microsoft Kinect sensor, the second one chooses points of interest from the 3D skeleton to characterize the gestures, and the last one proposes a neural connectionist model to perform the supervised classification of data. The experimental results show that our system can achieve above 93.3% recognition rate.

Keywords: ASD, artificial neural network, kinect, stereotypical motor movements

Procedia PDF Downloads 306

Authors: Naouel Zoghlami

Abstract:

Spoken-word recognition involves the simultaneous activation of potential word candidates which compete with each other for final correct recognition. In continuous speech, the activation-competition process gets more complicated due to speech reductions existing at word boundaries. Lexical processing is more difficult in L2 than in L1 because L2 listeners often lack phonetic, lexico-semantic, syntactic, and prosodic knowledge in the target language. In this study, we investigate the on-line lexical segmentation hypotheses that French listeners of L2 English form and then revise as subsequent perceptual evidence is revealed. Our purpose is to shed further light on the processes of L2 spoken-word recognition in context and better understand L2 listening difficulties through a comparison of skilled and unskilled reactions at the point where their working hypothesis is rejected. We use a variant of the gating experiment in which subjects transcribe an English sentence presented in increments of progressively greater duration. The spoken sentence was “And this amazing athlete has just broken another world record”, chosen mainly because it included common reductions and phonetic features in English, such as elision and assimilation. Our preliminary results show that there is an important difference in the manner in which proficient and less-proficient L2 listeners handle connected speech. Less-proficient listeners delay recognition of words as they wait for lexical and syntactic evidence to appear in the gates. Further statistical results are currently being undertaken.

Keywords: gating paradigm, spoken word recognition, online lexical segmentation, L2 listening

Procedia PDF Downloads 464

Authors: Vinay A. Sharma, Shiva Prasad H. C.

Abstract:

The initial stages of any project are often observed to be in a mixed set of conditions. Setting up the project is a tough task, but taking the initial decisions is rather not complex, as some of the critical factors are yet to be introduced into the scenario. These simple initial decisions potentially shape the timeline and subsequent events that might later be plotted on it. Proceeding towards the solution for a problem is the primary objective in the initial stages. The optimization in the solutions can come later, and hence, the resources deployed towards attaining the solution are higher than what they would have been in the optimized versions. A ‘logic’ that counters the problem is essentially the core of the desired solution. Thus, if the problem is solved, the deployment of resources has led to the required logic being attained. As the project proceeds along, the individuals working on the project face fresh challenges as a team and are better accustomed to their surroundings. The developed, optimized solutions are then considered for implementation, as the individuals are now experienced, and know better of the consequences and causes of possible failure, and thus integrate the adequate tolerances wherever required. Furthermore, as the team graduates in terms of strength, acquires prodigious knowledge, and begins its efficient transfer, the individuals in charge of the project along with the managers focus more on the optimized solutions rather than the traditional ones to minimize the required resources. Hence, as time progresses, the authorities prioritize attainment of the required logic, at a lower amount of dedicated resources. For empirical analysis of the stated theory, leaders and key figures in organizations are surveyed for their ideas on appropriate logic required for tackling a problem. Key-pointers spotted in successfully implemented solutions are noted from the analysis of the responses and a metric for measuring logic is developed. A graph is plotted with the quantifiable logic on the Y-axis, and the dedicated resources for the solutions to various problems on the X-axis. The dedicated resources are plotted over time, and hence the X-axis is also a measure of time. In the initial stages of the project, the graph is rather linear, as the required logic will be attained, but the consumed resources are also high. With time, the authorities begin focusing on optimized solutions, since the logic attained through them is higher, but the resources deployed are comparatively lower. Hence, the difference between consecutive plotted ‘resources’ reduces and as a result, the slope of the graph gradually increases. On an overview, the graph takes a parabolic shape (beginning on the origin), as with each resource investment, ideally, the difference keeps on decreasing, and the logic attained through the solution keeps increasing. Even if the resource investment is higher, the managers and authorities, ideally make sure that the investment is being made on a proportionally high logic for a larger problem, that is, ideally the slope of the graph increases with the plotting of each point.

Keywords: decision-making, leadership, logic, strategic management

Procedia PDF Downloads 108

Authors: Jiahe Liu, HongyangYu, Feng Qian, Miao Luo

Abstract:

This study aimed to address the problem of improving the confidence of key points by fusing multi-view information, thereby estimating human posture more accurately. We first obtained multi-view image information and then used the MvP algorithm to fuse this multi-view information together to obtain a set of high-confidence human key points. We used these as the input for the Spatio-Temporal Graph Convolution (ST-GCN). ST-GCN is a deep learning model used for processing spatio-temporal data, which can effectively capture spatio-temporal relationships in video sequences. By using the MvP algorithm to fuse multi-view information and inputting it into the spatio-temporal graph convolution model, this study provides an effective method to improve the accuracy of human posture estimation and provides strong support for further research and application in related fields.

Keywords: multi-view, pose estimation, ST-GCN, joint fusion

Procedia PDF Downloads 70

Authors: Tokihiko Akita, Seiichi Mita

Abstract:

A millimeter-wave radar is the most robust against adverse environments, making it an essential environment recognition sensor for automated driving. However, the reflection signal is sparse and unstable, so it is difficult to obtain the high recognition accuracy. Deep learning provides high accuracy even for them in recognition, but requires large scale datasets with ground truth. Specially, it takes a lot of cost to annotate for a millimeter-wave radar. For the solution, utilizing a simulator that can generate an annotated huge dataset is effective. Simulation of the radar is more difficult to match with real world data than camera image, and recognition by deep learning with higher-order features using the simulator causes further deviation. We have challenged to improve the accuracy of traffic participant classification by fusing simulator and real-world data with domain adaptation technique. Experimental results with the domain adaptation network created by us show that classification accuracy can be improved even with a few real-world data.

Keywords: millimeter-wave radar, object classification, deep learning, simulation, domain adaptation

Procedia PDF Downloads 93

Authors: Gamze Uslu, Sebnem Baydere, Alper K. Demir

Abstract:

In this study, data loss tolerance of Support Vector Machines (SVM) based activity recognition model and multi activity classification performance when data are received over a lossy wireless sensor network is examined. Initially, the classification algorithm we use is evaluated in terms of resilience to random data loss with 3D acceleration sensor data for sitting, lying, walking and standing actions. The results show that the proposed classification method can recognize these activities successfully despite high data loss. Secondly, the effect of differentiated quality of service performance on activity recognition success is measured with activity data acquired from a multi hop wireless sensor network, which introduces high data loss. The effect of number of nodes on the reliability and multi activity classification success is demonstrated in simulation environment. To the best of our knowledge, the effect of data loss in a wireless sensor network on activity detection success rate of an SVM based classification algorithm has not been studied before.

Keywords: activity recognition, support vector machines, acceleration sensor, wireless sensor networks, packet loss

Procedia PDF Downloads 475

Authors: Michele Aleandri, Matteo Colangeli, Davide Gabrielli

Abstract:

We study a generalization to a continuous setting of the classical Markov chain tree theorem. In particular, we consider an irreducible diffusion process on a metric graph. The unique invariant measure has an atomic component on the vertices and an absolutely continuous part on the edges. We show that the corresponding density at x can be represented by a normalized superposition of the weights associated to metric arborescences oriented toward the point x. A metric arborescence is a metric tree oriented towards its root. The weight of each oriented metric arborescence is obtained by the product of the exponential of integrals of the form ∫a/b², where b is the drift and σ² is the diffusion coefficient, along the oriented edges, for a weight for each node determined by the local orientation of the arborescence around the node and for the inverse of the diffusion coefficient at x. The metric arborescences are obtained by cutting the original metric graph along some edges.

Keywords: diffusion processes, metric graphs, invariant measure, reversibility

Procedia PDF Downloads 172

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: Van Trieu, Shouhuai Xu, Yusheng Feng

Abstract:

Tracking attack trajectories can be difficult, with limited information about the nature of the attack. Even more difficult as attack information is collected by Intrusion Detection Systems (IDSs) due to the current IDSs having some limitations in identifying malicious and anomalous traffic. Moreover, IDSs only point out the suspicious events but do not show how the events relate to each other or which event possibly cause the other event to happen. Because of this, it is important to investigate new methods capable of performing the tracking of attack trajectories task quickly with less attack information and dependency on IDSs, in order to prioritize actions during incident responses. This paper proposes a two-level graph causality framework for tracking attack trajectories in internet networks by leveraging observable malicious behaviors to detect what is the most probable attack events that can cause another event to occur in the system. Technically, given the time series of malicious events, the framework extracts events with useful features, such as attack time and port number, to apply to the conditional independent tests to detect the relationship between attack events. Using the academic datasets collected by IDSs, experimental results show that the framework can quickly detect the causal pairs that offer meaningful insights into the nature of the internet network, given only reasonable restrictions on network size and structure. Without the framework’s guidance, these insights would not be able to discover by the existing tools, such as IDSs. It would cost expert human analysts a significant time if possible. The computational results from the proposed two-level graph network model reveal the obvious pattern and trends. In fact, more than 85% of causal pairs have the average time difference between the causal and effect events in both computed and observed data within 5 minutes. This result can be used as a preventive measure against future attacks. Although the forecast may be short, from 0.24 seconds to 5 minutes, it is long enough to be used to design a prevention protocol to block those attacks.

Keywords: causality, multilevel graph, cyber-attacks, prediction

Procedia PDF Downloads 156

Authors: B. R. Campomanes-Alvarez, P. Quiros, B. Fernandez

Abstract:

Automatic Speech Recognition (ASR) is a machine-based process of decoding and transcribing oral speech. A typical ASR system receives acoustic input from a speaker or an audio file, analyzes it using algorithms, and produces an output in the form of a text. Some speech recognition systems use Hidden Markov Models (HMMs) to deal with the temporal variability of speech and Gaussian Mixture Models (GMMs) to determine how well each state of each HMM fits a short window of frames of coefficients that represents the acoustic input. Another way to evaluate the fit is to use a feed-forward neural network that takes several frames of coefficients as input and produces posterior probabilities over HMM states as output. Deep neural networks (DNNs) that have many hidden layers and are trained using new methods have been shown to outperform GMMs on a variety of speech recognition systems. Acoustic models for state-of-the-art ASR systems are usually training on massive amounts of data. However, audio files with their corresponding transcriptions can be difficult to obtain, especially in the Spanish language. Hence, in the case of these low-resource scenarios, building an ASR model is considered as a complex task due to the lack of labeled data, resulting in an under-trained system. Semi-supervised learning approaches arise as necessary tasks given the high cost of transcribing audio data. The main goal of this proposal is to develop a procedure based on acoustic semi-supervised learning for Spanish ASR systems by using DNNs. This semi-supervised learning approach consists of: (a) Training a seed ASR model with a DNN using a set of audios and their respective transcriptions. A DNN with a one-hidden-layer network was initialized; increasing the number of hidden layers in training, to a five. A refinement, which consisted of the weight matrix plus bias term and a Stochastic Gradient Descent (SGD) training were also performed. The objective function was the cross-entropy criterion. (b) Decoding/testing a set of unlabeled data with the obtained seed model. (c) Selecting a suitable subset of the validated data to retrain the seed model, thereby improving its performance on the target test set. To choose the most precise transcriptions, three confidence scores or metrics, regarding the lattice concept (based on the graph cost, the acoustic cost and a combination of both), was performed as selection technique. The performance of the ASR system will be calculated by means of the Word Error Rate (WER). The test dataset was renewed in order to extract the new transcriptions added to the training dataset. Some experiments were carried out in order to select the best ASR results. A comparison between a GMM-based model without retraining and the DNN proposed system was also made under the same conditions. Results showed that the semi-supervised ASR-model based on DNNs outperformed the GMM-model, in terms of WER, in all tested cases. The best result obtained an improvement of 6% relative WER. Hence, these promising results suggest that the proposed technique could be suitable for building ASR models in low-resource environments.

Keywords: automatic speech recognition, deep neural networks, machine learning, semi-supervised learning

Procedia PDF Downloads 339