research centers


Search results: Found 5

Listing 1 - 5 of 5
Sort by

Article
Current Big Data Issues and Their Solutions via Deep Learning: An Overview

Authors: Roohie Naaz Mir --- Asif Ali Banka
Journal: Iraqi Journal for Electrical And Electronic Engineering المجلة العراقية للهندسة الكهربائية والالكترونية ISSN: 18145892 Year: 2018 Volume: 14 Issue: 2 Pages: 127-138
Publisher: Basrah University جامعة البصرة

Loading...
Loading...
Abstract

The advancements in modern day computing and architectures focus on harnessing parallelism and achieve high performance computing resulting in generation of massive amounts of data. The information produced needs to be represented and analyzed to address various challenges in technology and business domains. Radical expansion and integration of digital devices, networking, data storage and computation systems are generating more data than ever. Data sets are massive and complex, hence traditional learning methods fail to rescue the researchers and have in turn resulted in adoption of machine learning techniques to provide possible solutions to mine the information hidden in unseen data. Interestingly, deep learning finds its place in big data applications. One of major advantages of deep learning is that it is not human engineered. In this paper, we look at various machine learning algorithms that have already been applied to big data related problems and have shown promising results. We also look at deep learning as a rescue and solution to big data issues that are not efficiently addressed using traditional methods. Deep learning is finding its place in most applications where we come across critical and dominating 5Vs of big data and is expected to perform better.


Article
Indoor Localization Using Deep-Learning and Smartphone

Authors: Zainab Mohammed Resan --- Muayad Sadik Croock
Journal: IRAQI JOURNAL OF COMPUTERS,COMMUNICATION AND CONTROL & SYSTEMS ENGINEERING المجلة العراقية لهندسة الحاسبات والاتصالات والسيطرة والنظم ISSN: 18119212 Year: 2019 Volume: 19 Issue: 3 Pages: 40-49
Publisher: University of Technology الجامعة التكنولوجية

Loading...
Loading...
Abstract

Robust and accurate indoor localization has been the goal of several research efforts over the past decade. In the building where the GPS is not available, this project can be utilized. Indoor localization based on image matching techniques related to deep learning was achieved in a hard environment. So, if it wanted to raise the precision of indoor classification, the number of image dataset of the indoor environment should be as large as possible to satisfy and cover the underlying area. In this work, a smartphone camera is used to build the image-based dataset of the investigated building. In addition, captured images in real time are taken to be processed with the proposed model as a test set. The proposed indoor localization includes two phases the first one is the offline learning phase and the second phase is the online processing phase. In the offline learning phase, here we propose a convolutional neural network (CNN) model that sequences the features of image data from some classis's dataset composed with a smartphone camera. In the online processing phase, an image is taken by the camera of a smartphone in real–time to be tested by the proposed model. The obtained results of the prediction can appoint the expected indoor location. The proposed system has been tested over various experiments and the obtained experimental results show that the accuracy of the prediction is almost 98.0%.


Article
Convolution Neural Networks for Blind Image Steganalysis: A Comprehensive Study

Authors: Hanaa Mohsin Ahmed --- Halah Hasan Mahmoud
Journal: Journal of Al-Qadisiyah for Computer Science and Mathematics مجلة القادسية لعلوم الحاسوب والرياضيات ISSN: 20740204 / 25213504 Year: 2019 Volume: 11 Issue: 2 Pages: Comp Page 53-64
Publisher: Al-Qadisiyah University جامعة القادسية

Loading...
Loading...
Abstract

Recently, Convolution Neural Network is widely applied in Image Classification, Object Detection, Scene labeling, Speech, Natural Language Processing and other fields. In this comprehensive study a variety of scenarios and efforts are surveyed since 2014 at yet, in order to provide a guide to further improve future researchers what CNN-based blind image steganalysis are presented its architecture, performance and limitations. Long-standing and important problem in image steganalysis difficulties mainly lie in how to give high accuracy and low payload in stego or cover images for improving performance of the network.


Article
Advantages and Disadvantages of Automatic Speaker Recognition Systems

Authors: Rawia Ab. Mohammed --- Akbas E. Ali --- Nidaa F. Hassan
Journal: Journal of Al-Qadisiyah for Computer Science and Mathematics مجلة القادسية لعلوم الحاسوب والرياضيات ISSN: 20740204 / 25213504 Year: 2019 Volume: 11 Issue: 3 Pages: Comp Page 21-30
Publisher: Al-Qadisiyah University جامعة القادسية

Loading...
Loading...
Abstract

Automatic speaker recognition systems use the machines to recognize an individual via a spoken sentence. Those systems recognize a specific individual or confirm an individual’s claimed identity. The most common type of voice biometrics is the Speaker Recognition. Its task focused on validation of a person’s claimed identity, using features that have been obtained via their voices. Throughout the last decades a wide range of new advances in the speaker recognition area have been accomplished, but there are still many problems that need solving or require enhanced solutions. In this paper, a brief overview of speech processing is given firstly, then some feature extraction and classifier techniques are described, also a comparative and analysis of some previous research are studied in depth, all this work leads to determine the best methods for speaker recognition. Adaptive MFCC and Deep Learning methods are determined to be more efficient and accurate than other methods in speaker recognition, thus these methods are recommend to be more suitable for practical applications.


Article
Quad-Copter Design and Fabrication by using Neural Network idea based on Advanced Microprocessor
تصميم وتصنيع طائرة رباعية بأستخدام الشبكة العصبية أستنادا الى معالج متقدم

Authors: Osama Qasim Jumah --- Haneen Safi Kadhim
Journal: journal of kerbala university مجلة جامعة كربلاء ISSN: 18130410 Year: 2018 Volume: 16 Issue: 2 Pages: 68-82
Publisher: Kerbala University جامعة كربلاء

Loading...
Loading...
Abstract

The current work aims to design and implement an independent quadcopter for locating a particular location and landing on a station of the required target. Where an outdoor quadcopter is designed and its flight is done by automatic flight. The quadcopter requires a wide control system for flight. Operation and tuning processes of the system become very difficult with existence of many parameters. Therefore, PID controller is optimized by the Invasive Weed Optimization (IWO) algorithm that is used to make quadcopter more stable, in addition to the sensors that help in achieving the stability and equilibrium for the quadcopter.In the present work, movements of the quadcopter that are named as roll, pitch and yaw are controlled by three PID controllers designed for this purpose. Here, the LattePanda controller board, USB camera that is connected with the board of the quadcopter and Neo-M8N GPS are used to locate the target and monitor the (X) mark during the automated landing of the quadcopter. Matlab 2014b program is installed inside the microprocessor LattePanda which can detect the object (Mark X) by using Deep Learning Algorithm. The control unit (PID) was easy to implement the simulation system by using the Matlab 2014b and required a short execution time during the simulation.

يهدف العمل الحالي الى تصميم وتنفيذ كوادكوبتر مستقلة لتحديد موقع معين وللهبوط على محطة الهدف المطلوب. حيث تم تصميم كوادكوبتر خارجي، يتم طيرانها أما عن طريق الطيران تلقائي أو عن طريق التحكم عن بعد. يتضمن التصميم على جميع الامور البرمجية والمادية والمسائل النظرية. الكوادكوبتر تتطلب نظام مراقبة واسع للطيران حيث عمليات تشغيل وضبط النظام تصبح جدا معقده مع وجود العديد من المكونات والمعاملات. لذلك تم استخدام وحده التحكم PID)) مع خوارزمية تحسين الأعشاب الضارة IWO)) التي تستخدم لجعل الكوادكوبتر اكثر استقرارا، بالإضافة الى أجهزة الاستشعار التي تساعد في تحقيق الاستقرار و التوازن للكوادكوبتر. في العمل الحالي، حركات الطائرة التي تدعى: زاوية الخطران ( Roll) و زاوية العطوف (Pitch) و زاوية الأنعراج (Yaw) والأرتفاع (Throttle) يتم التحكم بها عن طريق اربع وحدات تحكم نوع (PID) مصممة لهذا الغرض، وكذلك يتم أستخدام لوحة التحكم (LattePanda) وللمرة الاولى في تصميم الكوادكوبتر و مايكروسوفت كاميرا (USB camera )على متن الكوادكوبتر ونظام تحديد المواقع ( Neo-M8N GPS) لتحديد موقع الهدف ورصد الهدف المقصود (Mark X) خلال عملية الهبوط الالي للكوادكوبتر. خلال برنامج ماتلاب Matlab 2014b)) الذي تم تنصيبه داخل لوحة التحكم (Lattepanda) ، يمكن تحديد كل من الهدف المقصود ، مركز الهدف والمركز المرجعي من خلال الكشف عن الكائن باستخدام خوارزمية التعلم العميق (Deep Learning Algorithm)، وتعتمد هذه الخوارزمية على إيجاد نقاط مشتركة بين صورة الهدف التي تم تحديدها وصورة المرجع.

Listing 1 - 5 of 5
Sort by
Narrow your search

Resource type

article (5)


Language

English (5)


Year
From To Submit

2019 (3)

2018 (2)