NSU INSTITUTIONAL REPOSITORY
North South University Institutional Repository showcases the university's intellectual contributions, including journal articles, conference proceedings, theses, and more. Explore the latest research and advance your knowledge today!

Communities in NSU Repository
Select a community to browse its collections.
Confucius Institute at North South University North South University Journals North South University Library North South University Publications School of Business & Economics (SBE) School of Engineering & Physical Sciences (SEPS) School of Health & Life Sciences (SHLS) School of Humanities & Social Sciences (SHSS) The Office of External Affairs (OEA)
Recent Submissions
Item
Open Access
Sustainable Eco-friendly Reward Bin Using RVM by Detecting Recyclable waste and Generating points according to the weight of the Materials.
(North South University, 2019-09-30) Md. Rubaith Adnan; Mehedy Hassan; Md. Anowar Hosen; Mohammad Bin Khesru; Zunayeed Bin Zahir; 1510538642; 1521214042; 1512366043; 1321489042
In this project we present a system where the machine will receive three types of industrialized bottles and according to their weight it will generate some point for the user. With certain amount of points the user can collect money or some sort of reward depends on the system manufactures and the system service providers. First of all we will build a RVM (Reverse Vending Machine) and it can take three types of industrialized bottles like plastic bottles, Aluminum cans or Glass jars then it will measure the weight and pass to the next phase to sort it out. In the next phase the sensors will differentiate among these materials and will sort it and put them on desired container. While sorting the materials it already generated some points according to it’s weight. As we already set the threshold weight for the different materials the point Generation will be valid. The user will get a slip with point mentioned on it and s/he can collect The reward using that slip. The service provider will decide what type of reward a user will get At what amount of points. This project idea was developed because of the excessive amount of Plastic bottles on the street and it is harmful for the environment also. The recycling industries Are very profitable but it needs lots of manpower and it is not cost effective, so with this Machine it will be cost efficient as well as environment friendly. The people will encourage to Use the machine as they are benefiting. Thus this project is profitable from economic aspect at the same time it is very useful for keeping the environment from pollution which can not be reversed these days. We can see people using trash here and there even the vast ocean is getting polluted with this types of material everyday. No government or organization can do anything on their own in this regard but making people interested to keep the environment clean can be the next step to save the environment from destroying completely.
Item
Open Access
Reconstructing Better Audio by Using Spetial Microphone
(North South University, 2019-08-31) Md Mantasarul Elahi; Dr. Mohammad Ashrafuzzaman Khan; 1512518042
This article describes the design and implementation of a program that is capable of taking audio from multiple sources and using two filters, it will create the best possible version of the audio that is possible. Firstly, after taking in audio from multiple audio recording sources, it will firstly go through ’Unwanted Noise Reduction Filter’ and its product will be a noise reduced output. After going through the first filter, the audio files will be filtered once again using ‘Voice Enhancement Filter’ and will produce an enhanced output. Lastly, all the audio files will be combined using ‘Audio Merging Process’ to produce the final output of the best possible version of the audio that has been recorded
Item
Open Access
Chikitsa Nin: Real Time Web Based Remote Patient Management System using Internet of Things
(North South University, 2020) Md. Hasibur Rahman Fahim; Kho. Iftekhar Alam; Oyshee Rahaman; Hossain Ahmed Sojib; Tanjila Farah; 1612241042; 1510093042; 1611872042; 1610645042
Bangladesh currently offers six registered doctors for 10,000 patients, with a 43% vacancy in the required number of doctors. Here, the modernized hospitals are mostly urban area-based, and the scarcity of hospital accommodation is very real. Diagnostics are mostly private and out of reach for 24.3% of people living below the poverty line, and a hard call for almost 50% of the population who belong to the middle class. No professional medical consultancy is available here to find the best doctor in the field for emergencies, nor is there a hospital that provides the required service at once. Here, family members of critical patients, along with the patients, are often found running from hospital to hospital to check for the availability of their needed service. This suffering makes the “Chikitsa Nin” rise as a web application which creates a platform for its users to find a doctor to consult, get a medical representative to collect the sample for diagnosis, get required services from the hospitals around by choice, and medicine delivery altogether from one place.
Item
Open Access
Travel Buddy Finder
(North South University, 2019) Md.Farhan bin shafiq; Md.Shamsuzzaman; Md.Imam hossain; Mohammad Monirujjaman Khan; 1411464042; 0930037042; 1410242042
In this project, we develop and design an online-based application. Travel Buddy is a FREE social networking app that helps you find travel buddies and partners for your backpacking or holiday plans. We developed this application to automate a travel agency. This client-server-based app will serve all the agents of a travel agency and record all the information of a passenger. The agency can discover all the money transactions and passenger services from remote places through this platform. This social networking app also supplies necessary reports. In this project, there is a login page for each travel agency employee. The admin user can define roles for all users with respect to their user type. That means the admin can authenticate which kind of user can see or perform operations on which pages. This online-based app is device-independent.
Item
Open Access
Design and Implementation of a Blockchain Enabled E-Voting Application to Prevent Multiple Voting Schemes
(North South University, 2021) Iftekhar Mahmud Tahir; Md. Sifaat; Md. Imran Hasan; Mr. Zunayeed Bin Zahir; 1520541642; 1632285042; 1711460042
This report presents a Blockchain-based e-voting application that can prevent voters from casting their votes multiple times. The first and foremost step in this system is to validate the voter's identity. At first, the voter needs to download the e-voting application on their mobile, laptop, or other smart devices. After that, the voter needs to submit their identity information, which is verified by the organization conducting these elections. The administrator will check the registered voters' database and confirm whether they can vote. Then all the voter information will be securely added to the Blockchain. After the voter's identity is verified, a smart contract will be executed to issue a ballot to vote and submit it to the ballot box in the application. The application will ensure that a user cannot vote multiple times. When a user casts a vote, the administrator will check the Blockchain to ensure the voter has not already used up their vote. If the user's vote is valid, the administrator accepts their vote. In case the vote is found to be invalid, the administrator rejects their vote. Once a user casts a vote, it becomes a transaction and gets stored in the Blockchain encrypted. Moreover, the vote cannot be modified because of Blockchain's innate immutable characteristics. The voter will be given the option to print the receipt as proof of casting a vote. Through Blockchain, the voter will be able to verify that their vote has been cast and counted. The voter can even audit each ballot in the ballot box and confirm if the election results are accurate while retaining other voters' privacy.
Item
Open Access
OnDoctor-Telemedicine Service
(North South University, 2019-08-31) Rezaul Karim; Imtiaz Habib; Homiara Islam Parisa; Dr. Mohammad Monirujjaman Khan; 1410185042; 1420433042; 1420472042
This is a web based system using Django framework for backend development and for frontend it used html5, css3, bootstrap4, here patient can video call to registered doctor, chat with the doctor, booked appointment, buy medicine online, web application where user can find the primary health care by talking to the doctors online through video call system and also book appointment to the required doctor through this platform. The doctors are specialized and highly professional in their field. Website has also blog site where doctor post different health issue to make awareness among the society.
Item
Open Access
Plant Disease Detection Using Image Processing
(North South University, 2019-08-31) Sakif Obaid; Md. Sakib Iqbal; Md Jahidul Islam; Shamima Parvin; Mr. Zunayeed Bin Zahir; 1320297042; 1510423042; 1510301042; 1520321042
Agricultural productivity highly required on a global economy scale. This is the one of the reasons that disease detection in plants plays an important role in agriculture field, as having disease in plants are quite natural. Lack of care in this department will cause serious issues on plants for which respective product quality, quantity or productivity is affected. In this project, we have build an android application that can identify plant disease. Most of the plant have symptoms of any disease on its plant. Our application also has been developed based on dataset of plant leaf. This application will detect the disease on the plant and also suggest proper treatment for that disease. Various image processing and recognition technique has been used. Neural network has been introduced for getting satisfactory level of accuracy. Plant disease identification by visual way is more laborious task and at the same time, less accurate and can be done only in limited areas. Whereas if automatic detection technique is used it will take less effort, less time and become more accurate. In plants, some general diseases seen are brown and yellow spots, early and late scorch, and others are fungal, viral and bacterial diseases. Image processing is used for measuring affected area of disease and to determine the difference in the color of the affected area. Automatic detection of plant disease is beneficial as it reduces a huge task of monitoring in big farms of crops, can detect the symptoms of diseases at an early stage i.e. when they appear on plant leaves.
Item
Open Access
Eliya: An AI Robot, Assistant of the ECE Department, NSU
(North South University, 2019) Sakib Ahmed; Debashish Paul; Rubaiya Masnun; Minhaz Uddin Ahmed Shanto; Tanjila Farah; 1611100042; 1520935642; 1610273042; 1521144642
In our ECE Department at North South University, there are almost seven thousand students and hundreds of faculty members, many of them often have the same queries regarding departmental staff, but there are not enough human resources to answer all of them. Also, the officers assigned to the help desk at the department may sometimes get bored giving the same sort of answers again and again to a lot of people, and their office hours are limited, too. This is why we have come up with an idea to develop an AI Robot for our department, which will be able to give information regarding the departmental staff to the students and faculty members. Now, about Artificial Intelligence (AI)- Artificial Intelligence is one of the most invigorating fields in Robotics. There has always been a question about a robot’s intelligence, but nowadays Robotics is reaching a new level in the AI field. Usually, an AI Robot gathers information about an incident or situation through sensors or human inputs, and in our project, we use human inputs to make our robot useful for assistance. Our Robots will assist only in the fields it is programmed to assist in; they will not have the general analytical ability. But if our robot cannot help with something, it will respond by showing its inability to do that. Now, going to the features and how they are going to work, our Robot will be voice-controlled, where we will be using Machine Learning and NLP for controlling. Then, for movement and speaking, we will be using motors and speakers in the robot. It will give feedback on different questions using AI. We will be using voice recognition in the robot to recognize specific people. The robot will take orders from the supervisor again with voice recognition. It will interact with the general public using Machine Learning. The actual challenge in making an AI Robot is to understand how natural intelligence works, for example, in our project, it would be really tough for the robot to always move in the right direction as people can. But if we can provide it with a good mapping Algorithm, then its problems will be reduced. We have decided to work on AI Robots as we believe robots will gradually move out from the industrial and scientific worlds into our daily lives.
Item
Open Access
Smart Phone Control Robotic Arm
(North South University, 2024) Sarah Al Aiz; Md Ibnul Hasan Abir; Intisar Tahmid Naheen; 1831254045; 1821604643
In the recent years with increasing development of robotic arm and the wireless communications, the demand for a system that could easily connect devices for transfer of data over a long distance - without cables, grew stronger. This project is presenting the development of a smartphone control robot arm. A mobile robot that functional to do pick and place operation and be controlled by using Bluetooth H05 controller. It can move forward, reverse, turn right and left for a specific distance according to the controller specification. The development of this robot is based on Arduino UNO platform that will be interfaced with the smartphone application controller to the robotic arm. Analysis such as speed, distance, load that can be lifted of the robot has been done in order to know its performance. Finally, this prototype of the robot is expected to overcome the problem such as placing or picking object that far away from the user, pick and place hazardous object in the fastest and easiest way. This project will inspire the future generation for using the robotic arm and also informed the advantage of using this robot.
Item
Open Access
Smart Detection of Deepfake Using Various Deep Learning Models
(North South University, 2023) Mariam Binte Bashir; Iftekher Mahbub Rafi; Mohammad Monirujjaman Khan; 2021874042; 2021463642
Deepfake is a digitally manipulated image or video that is generated using an algorithm to replace the original person with someone else which looks authentic. The need to create efficient techniques for identifying and categorizing counterfeit photos on digital platforms has become crucial due to their widespread presence. This project does the prediction of counterfeit photos by employing of deep learning (DL) and transfer learning methodologies, specifically leveraging convolutional neural networks (CNNs) models. Our work focuses to enhance the accuracy and reliably detecting deep fake images. The DL models have been trained using the dataset acquired from Kaggle, titled "140k Real and Fake Images". Transfer learning techniques are employed to enhance prediction performance by leveraging knowledge from large datasets and applying it to pre-trained models. Evaluation criteria such as accuracy, precision, recall, and F1 score are employed to assess the ability of a model to effectively distinguish between genuine and manipulated images. The models exhibit strong discriminatory capabilities and provide reliable picture classifications. The study investigates the efficacy of various deep-learning models in the task of image classification. The Vgg16 achieved a peak accuracy of 99%, showcasing the possibilities of deep learning algorithms. MobileNetV3s and DenseNet121 demonstrate their efficacy in faulty image categorization by achieving accuracies of 98%, 94%, and 95% correspondingly. Unlike other models, Xception achieves remarkable accuracy rates of 96% through training on a wide range of datasets containing both genuine and manipulated images. This enables it to effectively differentiate between real image and deepfakes, which we deployed to build a website. Thus, our experiment demonstrates the versatility of deep learning models in handling various image formats.