Search results for: survey image overlaying

Authors: Khalide Jbilou

Abstract:

In the present paper, we propose numerical methods for solving the Stein equation AXC - X - D = 0 where the matrix A is large and sparse. Such problems appear in discrete-time control problems, filtering and image restoration. We consider the case where the matrix D is of full rank and the case where D is factored as a product of two matrices. The proposed methods are Krylov subspace methods based on the block Arnoldi algorithm. We give theoretical results and we report some numerical experiments.

Keywords: IEEEtran, journal, LATEX, paper, template.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1905

Authors: Fatema-Tuz-Zohra Khanam, Ali Al-Naji, Asanka G. Perera, Kim Gibson, Javaan Chahl

Abstract:

Conventional contact-based vital signs monitoring sensors such as pulse oximeters or electrocardiogram (ECG) may cause discomfort, skin damage, and infections, particularly in neonates with fragile, sensitive skin. Therefore, remote monitoring of the vital sign is desired in both clinical and non-clinical settings to overcome these issues. Camera-based vital signs monitoring is a recent technology for these applications with many positive attributes. However, there are still limited camera-based studies on neonates in a clinical setting. In this study, the heart rate (HR) and respiratory rate (RR) of eight infants at the Neonatal Intensive Care Unit (NICU) in Flinders Medical Centre were remotely monitored using a digital camera applying color and motion-based computational methods. The region-of-interest (ROI) was efficiently selected by incorporating an image decomposition method. Furthermore, spatial averaging, spectral analysis, band-pass filtering, and peak detection were also used to extract both HR and RR. The experimental results were validated with the ground truth data obtained from an ECG monitor and showed a strong correlation using the Pearson correlation coefficient (PCC) 0.9794 and 0.9412 for HR and RR, respectively. The root mean square errors (RMSE) between camera-based data and ECG data for HR and RR were 2.84 beats/min and 2.91 breaths/min, respectively. A Bland Altman analysis of the data also showed a close correlation between both data sets with a mean bias of 0.60 beats/min and 1 breath/min, and the lower and upper limit of agreement -4.9 to + 6.1 beats/min and -4.4 to +6.4 breaths/min for both HR and RR, respectively. Therefore, video camera imaging may replace conventional contact-based monitoring in NICU and has potential applications in other contexts such as home health monitoring.

Keywords: Neonates, NICU, digital camera, heart rate, respiratory rate, image decomposition.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 578

Authors: Mario Gonzalez, David Dominguez, Francisco B. Rodriguez

Abstract:

An attractor neural network on the small-world topology is studied. A learning pattern is presented to the network, then a stimulus carrying local information is applied to the neurons and the retrieval of block-like structure is investigated. A synaptic noise decreases the memory capability. The change of stability from local to global attractors is shown to depend on the long-range character of the network connectivity.

Keywords: Hebbian learning, image recognition, small world, spatial information.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1866

Authors: Mohammad Pooyan, Ali Taheri, Morteza Moazami-Goudarzi, Iman Saboori

Abstract:

In this paper we present a novel approach for wavelet compression of electrocardiogram (ECG) signals based on the set partitioning in hierarchical trees (SPIHT) coding algorithm. SPIHT algorithm has achieved prominent success in image compression. Here we use a modified version of SPIHT for one dimensional signals. We applied wavelet transform with SPIHT coding algorithm on different records of MIT-BIH database. The results show the high efficiency of this method in ECG compression.

Keywords: ECG compression, wavelet, SPIHT.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2398

Authors: Mohammad K. Abedin, Mohammad Shahjahan, Zeenat Sultana, Tawfique Jahan, Jesmin Akter

Abstract:

To strengthen social and behavior change communication (SBCC) capacity of Ministry of Health and Family Welfare (MoHFW) of the Government of Bangladesh, BCCP/BKMI developed two eLearning courses providing opportunities for professional development of SBCC Program Managers who have no access to training or refreshers training. The two eLearning courses – Message and Material Development (MMD) and Monitoring and Evaluation (MandE) of SBCC programs – went online in September 2015, where all users could register their participation so results could be monitored. Methodology: To assess the uses of these courses a randomly selected sample was collected to run a pre and post-test analyses and a phone survey were conducted. Systematic random sampling was used to select a sample of 75 MandE and 25 MMD course participants from a sampling frame of 179 and 51 respectively. Results: As of September 2016, more than 179 learners have completed the MandE course, and 49 learners have completed the MMD course. The users of these courses are program managers, university faculty members, and students. Encouraging results were revealed from the analysis of pre and post-test scores and a phone survey three months after course completion. Test scores suggested a substantial increase in knowledge. The pre-test scores findings suggested that about 19% learners scored high on the MandE. The post-test scores finding indicated a high score (92%) of the sample across 4 modules of MandE. For MMD course in pre-test scoring, 30% of the learners scored high, and 100% scored high at the post-test. It was found that all the learners in the phone survey have discussed the courses. Most of the sharing occurred with colleagues and friends, usually through face to face (70%) interaction. The learners reported that they did recommend the two courses to concerned people. About 67% MandE and 76% MMD learners stated that the concepts that they had to learn during the course were put into practice in their work settings. The respondents for both MandE and MMD courses have provided a valuable set of suggestions that would further strengthen the courses. Conclusions: The study showed that the initiative offered ample opportunities to build capacity in various ways in which the eLearning courses were used. It also highlighted the importance of scaling up these efforts to further strengthen the outcomes.

Keywords: E-learning course, message and material development, monitoring and evaluation, social and behavior change communication.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 867

Authors: L. Agneni, F. Carta, C. Micheletta, V. Tersigni

Abstract:

The development of a qualitative method to profile jobseekers is needed to improve the quality of the Public Employment Services (PES) in Italy. This is why the National Agency for Active Labour Market Policies (ANPAL) decided to introduce a Qualitative Profiling Service in the context of the activities carried out by local employment offices’ operators. The qualitative profiling service provides information and data regarding the jobseeker’s personal transition status, through a semi-structured questionnaire administered to PES clients during the guidance interview. The questionnaire responses allow PES staff to identify, for each client, proper activities and policy measures to support jobseekers in their reintegration into the labour market. Data and information gathered by the qualitative profiling tool are the following: frequency, modalities and motivations for clients to apply to local employment offices; clients’ expectations and skills; difficulties that they have faced during the previous working experiences; strategies, actions undertaken and activated channels for job search. These data are used to assess jobseekers’ personal and career characteristics and to measure their employability level (qualitative profiling index), in order to develop and deliver tailor-made action programmes for each client. This paper illustrates the use of the above-mentioned qualitative profiling service on the national territory and provides an overview of the main findings of the survey: concerning the difficulties that unemployed people face in finding a job and their perception of different aspects related to the transition in the labour market. The survey involved over 10.000 jobseekers registered with the PES. Most of them are beneficiaries of the “citizens' income”, a specific active labour policy and social inclusion measure. Furthermore, data analysis allows classifying jobseekers into a specific group of clients with similar features and behaviours, on the basis of socio-demographic variables, customers' expectations, needs and required skills for the profession for which they seek employment. Finally, the survey collects PES staff opinions and comments concerning clients’ difficulties in finding a new job and also their strengths. This is a starting point for PESs’ operators to define adequate strategies to facilitate jobseekers’ access or reintegration into the labour market.

Keywords: Labour market transition, Public Employment Services, qualitative profiling, vocational guidance.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 591

Authors: Yuji Kageyama, Shin Nagata, Tatsuya Takino, Izuru Nomura, Hiroyuki Kamata

Abstract:

In previous study, technique to estimate a self-location by using a lunar image is proposed.We consider the improvement of the conventional method in consideration of FPGA implementationin this paper. Specifically, we introduce Artificial Bee Colony algorithm for reduction of search time.In addition, we use fixed point arithmetic to enable high-speed operation on FPGA.

Keywords: SLIM, Artificial Bee Colony Algorithm, Location Estimate.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1980

Authors: Ebrahim Panah, Muhammad Yasir Babar

Abstract:

Thanks to the development of online technology and social networks, people are able to communicate as well as learn. WhatsApp is a popular social network which is growingly gaining popularity. This app can be used for communication as well as education. It can be used for instructor-learner, learner-learner, and learner-content interactions; however, very little knowledge is available on these potentials of WhatsApp. The current study was undertaken to investigate university students’ perceptions of WhatsApp used as a tool for instructor-learner dialogue, learner-content dialogue, and learner-learner dialogue. The study adopted a survey approach and distributed the questionnaire developed by Google Forms to 54 (11 males and 43 females) university students. The obtained data were analyzed using SPSS version 20. The result of data analysis indicates that students have positive attitudes towards WhatsApp as a tool for Instructor-Learner Dialogue: it easy to reach the lecturer (4.07), the instructor gives me valuable feedback on my assignment (4.02), the instructor is supportive during course discussion and offers continuous support with the class (4.00). Learner-Content Dialogue: WhatsApp allows me to academically engage with lecturers anytime, anywhere (4.00), it helps to send graphics such as pictures or charts directly to the students (3.98), it also provides out of class, extra learning materials and homework (3.96), and Learner-Learner Dialogue: WhatsApp is a good tool for sharing knowledge with others (4.09), WhatsApp allows me to academically engage with peers anytime, anywhere (4.07), and we can interact with others through the use of group discussion (4.02). It was also found that there are significant positive correlations between students’ perceptions of Instructor-Learner Dialogue (ILD), Learner-Content Dialogue (LCD), Learner-Learner Dialogue (LLD) and WhatsApp Application in classroom. The findings of the study have implications for lectures, policy makers and curriculum developers.

Keywords: Instructor-learner dialogue, learners-contents dialogue, learner-learner dialogue, WhatsApp.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 685

Authors: V. Tereshchenko, A. Tregubenko

Abstract:

In this study we survey the method for fast finding a minimum link path between two arbitrary points within a simple polygon, which can pass only through the vertices, with preprocessing.

Keywords: Minimum link path, simple polygon, Steiner points, optimal algorithm.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1511

Authors: Suprijanto, Linda Sari, Vebi Nadhira , IGN. Merthayasa. Farida I.M

Abstract:

Detection of human emotions has many potential applications. One of application is to quantify attentiveness audience in order evaluate acoustic quality in concern hall. The subjective audio preference that based on from audience is used. To obtain fairness evaluation of acoustic quality, the research proposed system for multimodal emotion detection; one modality based on brain signals that measured using electroencephalogram (EEG) and the second modality is sequences of facial images. In the experiment, an audio signal was customized which consist of normal and disorder sounds. Furthermore, an audio signal was played in order to stimulate positive/negative emotion feedback of volunteers. EEG signal from temporal lobes, i.e. T3 and T4 was used to measured brain response and sequence of facial image was used to monitoring facial expression during volunteer hearing audio signal. On EEG signal, feature was extracted from change information in brain wave, particularly in alpha and beta wave. Feature of facial expression was extracted based on analysis of motion images. We implement an advance optical flow method to detect the most active facial muscle form normal to other emotion expression that represented in vector flow maps. The reduce problem on detection of emotion state, vector flow maps are transformed into compass mapping that represents major directions and velocities of facial movement. The results showed that the power of beta wave is increasing when disorder sound stimulation was given, however for each volunteer was giving different emotion feedback. Based on features derived from facial face images, an optical flow compass mapping was promising to use as additional information to make decision about emotion feedback.

Keywords: Multimodal Emotion Detection, EEG, Facial Image, Optical Flow, compass mapping, Brain Wave

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2292

Authors: Sarisa Pinkham, Kanyarat Bussaban

Abstract:

The research aims to approximate the amount of daily rainfall by using a pixel value data approach. The daily rainfall maps from the Thailand Meteorological Department in period of time from January to December 2013 were the data used in this study. The results showed that this approach can approximate the amount of daily rainfall with RMSE=3.343.

Keywords: Daily rainfall, Image processing, Approximation, Pixel value data.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1758

Authors: A. Kumar, V. Devadas

Abstract:

Energy is the basic element for any country’s economic development. India is one of the most populated countries, and is dependent on fossil fuel and nuclear-based energy generation. The energy sector faces huge challenges and is dependent on the import of energy from neighboring countries to fulfill the gap in demand and supply. India has huge setbacks for efficient energy generation, distribution, and consumption, therefore they consume more quantity of energy to produce the same amount of Gross Domestic Product (GDP) compared to the developed countries. Technology and technique use, availability, and affordability in the various sectors are varying according to their economic status. In this paper, an attempt is made to quantify the domestic electrical energy consumption in Jodhpur, India. Survey research methods have been employed and stratified sampling technique-based households were chosen for conducting the investigation. Pre-tested survey schedules are used to investigate the grassroots level study. The collected data are analyzed by employing statistical techniques. Thereafter, a multiple regression model is developed to understand the functions of total electricity consumption in the domestic sector corresponding to other independent variables including electrical appliances, age of the building, household size, education, etc. The study resulted in identifying the governing variable in energy consumption at the household level and their relationship with the efficiency of household-based electrical and energy appliances. The analysis is concluded with the recommendation for optimizing the gap in peak electrical demand and supply in the domestic sector.

Keywords: Appliance, consumption, electricity, households.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 474

Authors: S. Malek, N. Zenati-Henda, M. Belhocine, S. Benbelkacem

Abstract:

In geometrical camera calibration, the objective is to determine a set of camera parameters that describe the mapping between 3D references coordinates and 2D image coordinates. In this paper, a technique of calibration and tracking based on both a least squares method is presented and a correlation technique developed as part of an augmented reality system. This approach is fast and it can be used for a real time system

Keywords: Camera calibration, pinhole model, least squares method, augmented reality, strong calibration.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2001

Authors: M.K. Bhuyan, Aragala Jagan.

Abstract:

Current image-based individual human recognition methods, such as fingerprints, face, or iris biometric modalities generally require a cooperative subject, views from certain aspects, and physical contact or close proximity. These methods cannot reliably recognize non-cooperating individuals at a distance in the real world under changing environmental conditions. Gait, which concerns recognizing individuals by the way they walk, is a relatively new biometric without these disadvantages. The inherent gait characteristic of an individual makes it irreplaceable and useful in visual surveillance. In this paper, an efficient gait recognition system for human identification by extracting two features namely width vector of the binary silhouette and the MPEG-7-based region-based shape descriptors is proposed. In the proposed method, foreground objects i.e., human and other moving objects are extracted by estimating background information by a Gaussian Mixture Model (GMM) and subsequently, median filtering operation is performed for removing noises in the background subtracted image. A moving target classification algorithm is used to separate human being (i.e., pedestrian) from other foreground objects (viz., vehicles). Shape and boundary information is used in the moving target classification algorithm. Subsequently, width vector of the outer contour of binary silhouette and the MPEG-7 Angular Radial Transform coefficients are taken as the feature vector. Next, the Principal Component Analysis (PCA) is applied to the selected feature vector to reduce its dimensionality. These extracted feature vectors are used to train an Hidden Markov Model (HMM) for identification of some individuals. The proposed system is evaluated using some gait sequences and the experimental results show the efficacy of the proposed algorithm.

Keywords: Gait Recognition, Gaussian Mixture Model, PrincipalComponent Analysis, MPEG-7 Angular Radial Transform.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1911

Authors: Juliana C. Alencar, Monica Ferreira do Amaral Porto

Abstract:

The blue-green infrastructure in recent years has been pointed out as a possibility to increase the environmental quality of watersheds. The regulation ecosystem services brought by these areas are many, such as the improvement of the air quality of the air, water, soil, microclimate, besides helping to control the peak flows and to promote the quality of life of the population. This study proposes a blue-green infrastructure scenario for the Jaguaré watershed, located in the western zone of the São Paulo city in Brazil. Based on the proposed scenario, it was verified the impact of the adoption of the blue and green infrastructure in the control of the peak flow of the basin, the benefits for the avifauna that are also reflected in the flora and finally, the quantification of the regulation ecosystem services brought by the adoption of the scenario proposed. A survey of existing green areas and potential areas for expansion and connection of these areas to form a network in the watershed was carried out. Based on this proposed new network of green areas, the peak flow for the proposed scenario was calculated with the help of software, ABC6. Finally, a survey of the ecosystem services contemplated in the proposed scenario was made. It was possible to conclude that the blue and green infrastructure would provide several regulation ecosystem services for the watershed, such as the control of the peak flow, the connection frame between the forest fragments that promoted the environmental enrichment of these fragments, improvement of the microclimate and the provision of leisure areas for the population.

Keywords: Blue and green infrastructure, sustainable drainage, urban waters, ecosystem services.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1001

Authors: Samson T. Obafemi, Seraphin D. Eyono Obono

Abstract:

Information and Communication Technologies (ICTs) are pervasive nowadays, including in education where they are expected to improve the performance of learners. However, the hope placed in ICTs to find viable solutions to the problem of poor academic performance in schools in the developing world has not yet yielded the expected benefits. This problem serves as a motivation to this study whose aim is to examine the perceptions of educators on the advantages and disadvantages of e-learning. This aim will be subdivided into two types of research objectives. Objectives on the identification and design of theories and models will be achieved using content analysis and literature review. However, the objective on the empirical testing of such theories and models will be achieved through the survey of educators from different schools in the Pinetown District of the South African Kwazulu-Natal province. SPSS is used to quantitatively analyse the data collected by the questionnaire of this survey using descriptive statistics and Pearson correlations after assessing the validity and the reliability of the data. The main hypothesis driving this study is that there is a relationship between the demographics of educators’ and their adherence to learning theories on one side, and their perceptions on the advantages and disadvantages of e-learning on the other side, as argued by existing research; but this research views these learning theories under three perspectives: educators’ adherence to self-regulated learning, to constructivism, and to progressivism. This hypothesis was fully confirmed by the empirical study except for the demographic factor where teachers’ level of education was found to be the only demographic factor affecting the perceptions of educators on the advantages and disadvantages of e-learning.

Keywords: Academic performance, e-learning, Learning theories, Teaching and Learning.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2634

Authors: Monika Chuchro, Maciej Dwornik

Abstract:

The extraction of meaningful information from image could be an alternative method for time series analysis. In this paper, we propose a graphical analysis of time series grouped into table with adjusted colour scale for numerical values. The advantages of this method are also discussed. The proposed method is easy to understand and is flexible to implement the standard methods of pattern recognition and verification, especially for noisy environmental data.

Keywords: graphical analysis, time series, seasonality, noisy environmental data

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1450

Authors: A. Melouah, H. Merouani

Abstract:

A new Markovianity approach is introduced in this paper. This approach reduces the response time of classic Markov Random Fields approach. First, one region is determinated by a clustering technique. Then, this region is excluded from the study. The remaining pixel form the study zone and they are selected for a Markovianity segmentation task. With Selective Markovianity approach, segmentation process is faster than classic one.

Keywords: Markovianity, response time, segmentation, study zone.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1458

Authors: Ali Shojaee Bakhtiari, Ali Asghar Beheshti Shirazi, Amir Sepasi Zahmati

Abstract:

An efficient iris segmentation method based on analyzing the local entropy characteristic of the iris image, is proposed in this paper and the strength and weaknesses of the method are analyzed for practical purposes. The method shows special strength in providing designers with an adequate degree of freedom in choosing the proper sections of the iris for their application purposes.

Keywords: Iris segmentation, entropy, biocryptosystem, biometric identification.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1428

Authors: Hritwik Ghosh, Irfan Sadiq Rahat, Sachi Nandan Mohanty, J. V. R. Ravindra, Abdus Sobur

Abstract:

In the rapidly evolving landscape of medical diagnostics, the early detection and accurate classification of skin cancer remain paramount for effective treatment outcomes. This research delves into the transformative potential of artificial intelligence (AI), specifically deep learning (DL), as a tool for discerning and categorizing various skin conditions. Utilizing a diverse dataset of 3,000 images, representing nine distinct skin conditions, we confront the inherent challenge of class imbalance. This imbalance, where conditions like melanomas are over-represented, is addressed by incorporating class weights during the model training phase, ensuring an equitable representation of all conditions in the learning process. Our approach presents a hybrid model, amalgamating the strengths of two renowned convolutional neural networks (CNNs), VGG16 and ResNet50. These networks, pre-trained on the ImageNet dataset, are adept at extracting intricate features from images. By synergizing these models, our research aims to capture a holistic set of features, thereby bolstering classification performance. Preliminary findings underscore the hybrid model's superiority over individual models, showcasing its prowess in feature extraction and classification. Moreover, the research emphasizes the significance of rigorous data pre-processing, including image resizing, color normalization, and segmentation, in ensuring data quality and model reliability. In essence, this study illuminates the promising role of AI and DL in revolutionizing skin cancer diagnostics, offering insights into its potential applications in broader medical domains.

Keywords: Artificial intelligence, machine learning, deep learning, skin cancer, dermatology, convolutional neural networks, image classification, computer vision, healthcare technology, cancer detection, medical imaging.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1463

Authors: Justina I. R. Udotong, Ime R. Udotong, Offiong U. Eka

Abstract:

The concentrations of heavy metals in sediments of Qua Iboe River Estuary (QIRE) were monitored at four different sampling locations in wet and dry seasons. A preliminary survey to determine the four sampling stations along the river continuum showed that the area spanned between <0.1‰ salinity at the control station and 21.5‰ at the fourth station along the river continuum. A preliminary survey to determine the four sampling locations along the river estuary showed variations in salinity and other physicochemical parameters. The estuary was found to be polluted with heavy metals from point and nonpoint sources at varying degrees. Mean values of 7.80 mg/kg, 4.97 mg/kg and 2.80 mg/kg of nickel were obtained for sediment samples from Douglas creek, Qua Iboe and Atlantic sampling locations, respectively in the dry season. The wet season nickel concentrations were however lower. The entire study area was grossly contaminated by iron. At Douglas creek, the concentration of iron in sediment was 9274 ± 9.54mg/kg while copper, nickel, lead and vanadium were <0.5mg/kg each as compared to iron. Bioaccumulation was therefore suspected within the study area as values of 31.00 ± 0.79, 36.00 ± 0.10 and 55.00 ± 0.05 mg/kg of zinc were recorded in sediment at Douglas creek, Atlantic and the control sampling locations. The results from this study showed that the source of these heavy metals were from point sources like the corrosion of metal steel pipes from old bridges as well as oily sludge wastes from the Qua Iboe Terminal / tank farm located within the vicinity of the study area.

Keywords: Heavy metal, Qua Iboe River Estuary, seasonal variations, sediment.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2122

Authors: Wei Zhang

Abstract:

With the rapid development of deep learning, neural network and deep learning algorithms play a significant role in various practical applications. Due to the high accuracy and good performance, Convolutional Neural Networks (CNNs) especially have become a research hot spot in the past few years. However, the size of the networks becomes increasingly large scale due to the demands of the practical applications, which poses a significant challenge to construct a high-performance implementation of deep learning neural networks. Meanwhile, many of these application scenarios also have strict requirements on the performance and low-power consumption of hardware devices. Therefore, it is particularly critical to choose a moderate computing platform for hardware acceleration of CNNs. This article aimed to survey the recent advance in Field Programmable Gate Array (FPGA)-based acceleration of CNNs. Various designs and implementations of the accelerator based on FPGA under different devices and network models are overviewed, and the versions of Graphic Processing Units (GPUs), Application Specific Integrated Circuits (ASICs) and Digital Signal Processors (DSPs) are compared to present our own critical analysis and comments. Finally, we give a discussion on different perspectives of these acceleration and optimization methods on FPGA platforms to further explore the opportunities and challenges for future research. More helpfully, we give a prospect for future development of the FPGA-based accelerator.

Keywords: Deep learning, field programmable gate array, FPGA, hardware acceleration, convolutional neural networks, CNN.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 895

Authors: A. Kebbeb, M. Mostefai, F. Benmerzoug, Y. Chahir

Abstract:

The multimodal biometric identification is the combination of several biometric systems; the challenge of this combination is to reduce some limitations of systems based on a single modality while significantly improving performance. In this paper, we propose a new approach to the construction and the protection of a multimodal biometric database dedicated to an identification system. We use a topological watermarking to hide the relation between face image and the registered descriptors extracted from other modalities of the same person for more secure user identification.

Keywords: Biometric databases, Multimodal biometrics, security authentication, Digital watermarking.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2090

Authors: Hazem M. El-Bakry

Abstract:

Principal Component Analysis (PCA) has many different important applications especially in pattern detection such as face detection / recognition. Therefore, for real time applications, the response time is required to be as small as possible. In this paper, new implementation of PCA for fast face detection is presented. Such new implementation is designed based on cross correlation in the frequency domain between the input image and eigenvectors (weights). Simulation results show that the proposed implementation of PCA is faster than conventional one.

Keywords: Fast Face Detection, PCA, Cross Correlation, Frequency Domain

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1797

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 APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1309

Authors: Hira Lal Gope, Hidekazu Fukai

Abstract:

The aim of this study is to develop a system which can identify and sort peaberries automatically at low cost for coffee producers in developing countries. In this paper, the focus is on the classification of peaberries and normal coffee beans using image processing and machine learning techniques. The peaberry is not bad and not a normal bean. The peaberry is born in an only single seed, relatively round seed from a coffee cherry instead of the usual flat-sided pair of beans. It has another value and flavor. To make the taste of the coffee better, it is necessary to separate the peaberry and normal bean before green coffee beans roasting. Otherwise, the taste of total beans will be mixed, and it will be bad. In roaster procedure time, all the beans shape, size, and weight must be unique; otherwise, the larger bean will take more time for roasting inside. The peaberry has a different size and different shape even though they have the same weight as normal beans. The peaberry roasts slower than other normal beans. Therefore, neither technique provides a good option to select the peaberries. Defect beans, e.g., sour, broken, black, and fade bean, are easy to check and pick up manually by hand. On the other hand, the peaberry pick up is very difficult even for trained specialists because the shape and color of the peaberry are similar to normal beans. In this study, we use image processing and machine learning techniques to discriminate the normal and peaberry bean as a part of the sorting system. As the first step, we applied Deep Convolutional Neural Networks (CNN) and Support Vector Machine (SVM) as machine learning techniques to discriminate the peaberry and normal bean. As a result, better performance was obtained with CNN than with SVM for the discrimination of the peaberry. The trained artificial neural network with high performance CPU and GPU in this work will be simply installed into the inexpensive and low in calculation Raspberry Pi system. We assume that this system will be used in under developed countries. The study evaluates and compares the feasibility of the methods in terms of accuracy of classification and processing speed.

Keywords: Convolutional neural networks, coffee bean, peaberry, sorting, support vector machine.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1554

Authors: Duygu Aydın, Aşina Gülerarslan, Süleyman Karaçor, Tarık Doğan

Abstract:

Sharing motivations of viral advertisements by consumers and the impacts of these advertisements on the perceptions for brand will be questioned in this study. Three fundamental questions are answered in the study. These are advertisement watching and sharing motivations of individuals, criteria of liking viral advertisement and the impact of individual attitudes for viral advertisement on brand perception respectively. This study will be carried out via a viral advertisement which was practiced in Turkey. The data will be collected by survey method and the sample of the study consists of individuals who experienced the practice of sample advertisement. Data will be collected by online survey method and will be analyzed by using SPSS statistical package program. Recently traditional advertisement mind have been changing. New advertising approaches which have significant impacts on consumers have been argued. Viral advertising is a modernist advertisement mind which offers significant advantages to brands apart from traditional advertising channels such as television, radio and magazines. Viral advertising also known as Electronic Word-of- Mouth (eWOM) consists of free spread of convincing messages sent by brands among interpersonal communication. When compared to the traditional advertising, a more provocative thematic approach is argued. The foundation of this approach is to create advertisements that are worth sharing with others by consumers. When that fact is taken into consideration, in a manner of speaking it can also be stated that viral advertising is media engineering. The content worth sharing makes people being a volunteer spokesman of a brand and strengthens the emotional bonds among brand and consumer. Especially for some sectors in countries which are having traditional advertising channel limitations, viral advertising creates vital advantages.

Keywords: Viral advertising, marketing, consumers, brands.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 2152

Authors: Sharifah Sumayyah Engku Alwi, Masrah Azrifah Azmi Murad

Abstract:

The development of the information technology and Internet has been transforming the healthcare industry. The internet is continuously accessed to seek for health information and there are variety of sources, including search engines, health websites, and social networking sites. Providing more and better information on health may empower individuals, however, ensuring a high quality and trusted health information could pose a challenge. Moreover, there is an ever-increasing amount of information available, but they are not necessarily accurate and up to date. Thus, this paper aims to provide an insight of the models and frameworks related to online health information seeking of consumers. It begins by exploring the definition of information behavior and information seeking to provide a better understanding of the concept of information seeking. In this study, critical factors such as performance expectancy, effort expectancy, and social influence will be studied in relation to the value of seeking health information. It also aims to analyze the effect of age, gender, and health status as the moderator on the factors that influence online health information seeking, i.e. trust and information quality. A preliminary survey will be carried out among the health professionals to clarify the research problems which exist in the real world, at the same time producing a conceptual framework. A final survey will be distributed to five states of Malaysia, to solicit the feedback on the framework. Data will be analyzed using SPSS and SmartPLS 3.0 analysis tools. It is hoped that at the end of this study, a novel framework that can improve online health information seeking is developed. Finally, this paper concludes with some suggestions on the models and frameworks that could improve online health information seeking.

Keywords: Information behavior, information seeking, online health information, technology acceptance model, the theory of planned behavior, UTAUT.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1697

Authors: Raymond Xu, Ashley Hua, Andrew Wang, Yuru Lin

Abstract:

During COVID-19, the depression rate has increased dramatically. Young adults are most vulnerable to the mental health effects of the pandemic. Lower-income families have a higher ratio to be diagnosed with depression than the general population, but less access to clinics. This research aims to achieve early depression detection at low cost, large scale, and high accuracy with an interdisciplinary approach by incorporating clinical practices defined by American Psychiatric Association (APA) as well as multimodal AI framework. The proposed approach detected the nine depression symptoms with Natural Language Processing sentiment analysis and a symptom-based Lexicon uniquely designed for young adults. The experiments were conducted on the multimedia survey results from adolescents and young adults and unbiased Twitter communications. The result was further aggregated with the facial emotional cues analyzed by the Convolutional Neural Network on the multimedia survey videos. Five experiments each conducted on 10k data entries reached consistent results with an average accuracy of 88.31%, higher than the existing natural language analysis models. This approach can reach 300+ million daily active Twitter users and is highly accessible by low-income populations to promote early depression detection to raise awareness in adolescents and young adults and reveal complementary cues to assist clinical depression diagnosis.

Keywords: Artificial intelligence, depression detection, facial emotion recognition, natural language processing, mental disorder.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1180

Authors: Luciano Nieddu, Giuseppe Manfredi

Abstract:

In this paper a Pattern Recognition algorithm based on a constrained version of the k-means clustering algorithm will be presented. The proposed algorithm is a non parametric supervised statistical pattern recognition algorithm, i.e. it works under very mild assumptions on the dataset. The performance of the algorithm will be tested, togheter with a feature extraction technique that captures the information on the closed two-dimensional contour of an image, on images of industrial mineral ores.

Keywords: K-means, Industrial ores classification, Invariant Features, Supervised Classification.

Procedia APA BibTeX Chicago EndNote Harvard JSON MLA RIS XML ISO 690 PDF Downloads 1381