Browsing by Author "Md. Shahriar Hussain"

Now showing 1 - 14 of 14
  • Thumbnail Image
    Item
    Open Access
    A Machine Learning and Embedded Approach to Detect Fire, Identify the Cause of it and Extinguish Using the Best Possible Extinguisher
    (North South University, 2023-02-01) Md. Rashiqur Rahman; Sabbir Ahmed; Md Shafin Islam Rudro; Sadia Aktar; Md. Shahriar Hussain; 1812366042; 1813515042; 1821054642; 1821516042
    Chemical fires are a major cause of devastating situations and fatalities in Bangladesh, where inadequate knowledge of extinguishing such fires has led to numerous deaths, including those of firefighters. To address this issue, we propose a machine learning and embedded systems approach to design a dynamic and portable prototype that can remotely monitor the fire incident and determine its cause. The prototype uses a dashboard camera to stream video and gas sensors to detect the chemical gas information in the environment to identify and extinguish the fire. The embedded approach clarifies the cause of the fire and suggests the most suitable extinguisher to the user. The device is economical, portable, user-friendly, and marketable in Bangladesh. The system produces satisfactory results with high accuracy in fire detection and maintains its objective of being low-cost and user-friendly. The device can be remotely controlled using a mobile phone and web-based application. The system's machine learning approach uses appropriate datasets and algorithms (YOLOV5), while the embedded approach employs suitable sensors and a microcontroller (ESP32). The system's accuracy has been shown to be perfect.
  • Thumbnail Image
    Item
    Embargo
    An Intelligent Traffic Control System for Trouble-free Movement of Ambulances and Fire Trucks on Heavy Traffic Roads in Bangladesh Using YOLOv5 Algorithm and GSM Module
    (North South University, 2022) TASFIA SAMIN; FARZANA HAQUE; NESHAT ANJUMAN MOURI; Md. Shahriar Hussain; 1821823042; 1821386042; 1812749042
    One of the biggest problems developing countries like Bangladesh are facing is traffic problems. Especially metropolitan cities in a country need a proper traffic control system for stress-free movement, so the situation gets less stressful. This work has been designed to create a smooth- movement process for emergency vehicles like ambulances and fire trucks, which needs to be accessible as soon as possible during traffic congestion. The system would notify the traffic controller after detecting the emergency vehicle among all the stuck vehicles. This work uses a dataset constructed and labelled manually. To train the model, we have used the YOLOv5 algorithm. The accuracy of our model is 94%. The model detects emergency vehicles like ambulances and fire trucks as well as other vehicles captured by the CCTV camera of the road. However, only the detection of emergency vehicles would be notified to the traffic controller through Arduino Uno and GSM Module. That way, the traffic controller could clear that road for the easy movement of those emergency vehicles.
  • Thumbnail Image
    Item
    Open Access
    Automated Fisheries Water Quality Monitoring System
    (North South University, 2020-04-30) Tanbin Akter Mitaly; Paplu Dash; Md. Shahriar Hussain; 1712438642; 1611420043
    Fisheries is one of the main prospects to grow economically also we benefit from it by the production of food and the nutrition that is required us to live by. For fisheries, it's important to maintain good water quality as it is an important factor that affects fish health and performance in the aquaculture production system. It matters what type of fish we are culturing in our system due to different fish species need a different range of water quality aspects in which they can survive. Different fish species have a different optimized point to growing well. Don't maintain the water quality can result in parasite infestations, poor growth, erratic behaviour and various disease system. The water quality is dependent on climate, seasonal changes also how a culture system is used. To maintain good water quality conditions we need to look at physical and chemical components of water that have an impact on water quality.
  • Thumbnail Image
    Item
    Open Access
    Blind Person Assistant - Object detection With Voice Feedback
    (North South University, 2022) Mosarrat Shazia Kabir, Syeda Karishma Naaz, Simon Uddin, Sanjida Akter,; Md. Shahriar Hussain
    This project was specially implemented to help visually impaired people. The project can detect different objects in real time, counts them, and delivers voice feedback. Since our project is an object detection system, the first thing we needed was a proper dataset. For this project, we used two datasets. One is from the Microsoft COCO object detection dataset, and another is from the Kaggle object detection dataset. However, the Kaggle dataset was not preprocessed. Therefore, we had to resize the image, and along with that, we had to make all the images in the same color format. To process the images, we used "OpenCV" as an image processing tool. Next, we moved to set the algorithm to train and test our system. We chose to use two commonly used algorithms. One is the SSD Mobilenet algorithm, and the other is the YOLO Algorithm. In our project, we used SSD Mobilenet v2, YOLO v4, and YOLO v7, which are one of the fastest and most accurate object detectors. The accuracy we got for SSD Mobilenet v2, YOLO v4, and YOLO v7 was 94%, 94%, and 98%. Among the three models, YOLO v7 had the highest accuracy. Therefore we chose YOLO v7 for further work. Then we implemented GTTS (Google Text To Speech) and a counting function that counts the object classes. Eventually, the system counts the objects in real-time detection, and with the help of GTTS, it returns voice feedback.
  • Thumbnail Image
    Item
    Open Access
    Decentralised Banking System Using Blockchain Technology and Solidity Contracts
    (North South University, 2023) Al Sadat Rafi; Md Riaz Uddin Mojumdar; Md. Mahir Asif Chowdhury; Md. Shahriar Hussain; 1813118642; 1912430042; 1712447642
    Decentralized Banking System Using Blockchain Technology and Solidity Contracts : This thesis aims to design and implement a decentralized banking system using blockchain technology and smart contracts. The objective of this study is to address the issues of security and transparency in traditional banking systems by utilizing the immutability and decentralization properties of blockchain technology. The proposed system will be built on the Ethereum blockchain platform and will use smart contracts written in Solidity to facilitate transactions and maintain a secure and transparent ledger of all banking activities. The system will also be evaluated for its scalability and potential real-world applications. The blockchain technology offers a secure, decentralized, and transparent way of recording transactions, which is well-suited for the banking industry. This decentralized banking system will enable users to have full control over their assets and transactions, without the need for intermediaries. The study will also explore the potential of this system in the context of financial inclusion and the challenges that need to be addressed for its successful implementation. The results of this study will provide valuable insights for the future development of decentralized banking systems and the integration of blockchain technology in the financial industry. The proposed system has the potential to revolutionize the traditional banking system by making it more secure, transparent, and efficient.
  • Thumbnail Image
    Item
    Open Access
    Line Following Automated Car Detecting Sign
    (North South University, 2023) Shadman Haque; Maheenul Hoque Chowdhury; SM Monjurul Alam; Md. Shahriar Hussain; 1911707642; 1921133642; 1921337042
    The advent of autonomous vehicles has transformed the transportation sector by offering more accessibility, reduced energy consumption, and improved safety. The ESP32-CAM module, which is equipped with a built-in camera, possesses a useful attribute: robustness. The module will be employed to construct a model of an autonomous car for the project's operation. The objective of the project is to provide dependable navigation and obstacle detection in many scenarios by leveraging computer vision techniques, cloud connectivity, and edge processing capabilities in a mutually beneficial manner. These technological innovations will enable us to accomplish our objective. The CPU is responsible for processing data from sensors, controlling motors, capturing real-time images, and enabling wireless connectivity. The algorithms developed on the microcontroller enable the identification and classification of elements in the vehicle's environment, such as lanes, barriers, and pedestrians. This is achieved by examining the frames captured by the camera. This study establishes the foundation for developing a versatile and user-friendly automated vehicle prototype with the microcontroller. It could potentially stimulate further research in this captivating domain and potentially contribute to the progress of technologies for autonomous vehicles. This is due to its integration of advanced technologies such as cloud connectivity, computer vision, and edge process.
  • Thumbnail Image
    Item
    Open Access
    Lip Reading App
    (North South University, 2023) Jubair Mahmud Pulock; Md. Shahriar Hussain; 1931111042
    Effective communication is essential for promoting inclusivity and closing the gap between people who speak different languages in today's globally connected world. But hearing-impaired people frequently have trouble understanding spoken language and must rely significantly on visual signals to follow discussions. To solve this problem, we introduce a revolutionary Lip Reading App, a state-of-the-art mobile application that uses computer vision and artificial intelligence (AI) to enable real-time speech interpretation. In order to effectively monitor and interpret lip movements and translate them into understandable text or audio output, the Lip Reading App makes use of the sophisticated capabilities of deep learning algorithms. The software recognizes and tracks facial landmarks using facial recognition technology, capturing the delicate movements and gestures involved in speech creation. The software can produce accurate transcriptions or audio renderings of the spoken content by processing these acquired visual cues and comparing them to a huge library of phonetic and linguistic patterns. This project presents the development and training of a character-level sequence-to sequence language model using recurrent neural networks (RNNs). The primary objective of the model is to generate text predictions given input sequences of characters. The model employs the 'binary_crossentropy' loss function, although it might be more suitable to use categorical cross-entropy for multi-class classification tasks. Training is performed over multiple epochs, with a learning rate of 1.0e-4 (0.0001) and an average epoch time of 35026 seconds. The model's progress is evaluated based on the loss value, which steadily decreases with each epoch. By the end of 100 epochs, the average loss stands at 63.765. Despite the potential improvements in the loss function selection, the model exhibits promising capabilities in generating coherent and meaningful text predictions. Further fine-tuning and optimization could potentially enhance the performance and versatility of the language model for various text generation tasks.
  • Thumbnail Image
    Item
    Open Access
    Machine Learning Powered Stroke Predicting Web Application
    (North South University, 2021) Eshan Ahmed; Redwan Ahmed; Shahriar Hossain; Md. Shahriar Hussain; 1712918642; 1721279042; 1712852642
    Stroke is one of the major causes of severe health issues and death cases in the world. A chance of strokes may develop gradually in one person's body. Early prediction of stroke is useful for prevention, so it is possible to reduce the chance of stroke by taking regular checkups. The main objective of our project is to detect the possibility of stroke by machine learning which is a subfield of artificial intelligence. This machine learning part will be a significant part of our work. Another important component of our work is that we built a web platform based on machine learning. Here users will be able to make predictions simply by putting some values on our stroke prediction form. Again, the required values are user-friendly, so one will not need to go through any unnecessary hassle to collect them. All of the required attributes are related to one's lifestyle and overall general health condition. After getting the user’s required attributes’ value, our system will analyze those by our machine learning model, which we built earlier. Then our system will send a response if the user has any chance of having a stroke or not. Among all attributes or features, some crucial attributes play a significant role in having a stroke in the future. Our system will also prompt the user to predict which aspects of their health they should pay attention to. Our system will also store users' input attributes and their corresponding prediction results. This will help our users to keep track of their health conditions. We built and tested our whole system with utmost precision so that it can help our users to predict stroke before it happens.
  • Thumbnail Image
    Item
    Open Access
    Medicine Delivery Drone
    (North South University, 2023) Firoz Maisha; Rimi Sadia Islam; Md. Shahriar Hussain; 1911340042; 1821938042
    Unmanned Aerial Vehicle (UAV) is becoming new generation revolution system in modern World. This is considered as ‘Drone’ technology using in various kind of work. In this report we are going to propose this sort of technology titled as ‘Medicine Delivery Drone’ for using multipurpose progress and development in helping regards to traffic management system. The motivation for this project is to reduce operational cost of any sort of short distance distribution of medical equipment. Also addressing the emergencies in disaster time, this project is going to be an essential and integral part in remote area where medicine responses are more likely a important things. Moreover, this technology will make the delivery of medicine more efficient and precise in short distance place where it needed the most. This project has been decorated with Hardware and Software Part. In Hardware part we construct this Drone by APM flight controller, DJI 2212 motors, DJI self-lock propeller, APM flight controller GPS etc. This project will illustrate to the solution of light weight medicine distribution to those places where reaching is very difficult. Undermining some future prospect, this project will further help us to reduce traffic jam related problem and this will be another well-being for the patients and people who frequently need medication in shorter period of time in place of unavailability of those medicine. In the software part, we made a website named "Medi-fly". The user of this website can use it for selecting their required medicine and share their location. After that, we may deliver their required medicine via our customized drone. In addition, the drone has a censored box that is used to keep the medicine and eject after arrival to the customer. For location we are using "Mission Planner" software.
  • Thumbnail Image
    Item
    Open Access
    Object Detecting Robot Identifying & Picking Up Objects
    (North South University, 2023) Raisa Akhtar; Zarin Musharrat; Alifa Khan; Md. Shahriar Hussain; 2012010042; 1931826642; 1931829642
    This research focuses on the development of an accurate object detection algorithm to aid robots in identifying and picking up objects in cluttered environments. Object detection is crucial for robotic control systems, particularly in industrial automation. Traditional object detection algorithms struggle with objects lacking texture, leading to the adoption of deep learning methods like YOLO (You Only Look Once) for improved performance. The study aims to enhance object localization accuracy, especially in cluttered scenes, to facilitate robot grasping tasks. While basic YOLO models can detect objects, precise localization remains challenging. The research proposes a solution using deep YOLOs for semantic segmentation, which significantly improves object localization accuracy. Despite being time-consuming, YOLOv8 proves highly efficient, achieving an accuracy of 84.7%. The ultimate goal is to implement a robot capable of detecting and identifying five objects (ball, bottle, car, cup, spoon), and then picking and placing them according to user instructions. For the hardware part, we have used the Arduino UNO microcontroller, which is quite budget-friendly. As this is just a small test project, only a robotic arm gripper is used here, so that it can only grab it if it recognizes the object.
  • Thumbnail Image
    Item
    Open Access
    Parkinson’s disease onset detection Using Machine Learning.
    (North South University, 2022) MD Samiul Hasan; Kazal Roy Sagar; Anika Ferdausy; Md. Shahriar Hussain; 1512466042; 1430784642; 1512232042
    Accurately identifying Parkinson’s disease (PD) at an early stage is unquestionably critical for halting its progression and giving patients access to disease-modifying medication. To achieve that goal, the premotor stage of Parkinson’s disease should be closely studied. Based on premotor traits, a unique deep-learning technique is introduced to determine if an individual has Parkinson’s disease or not. In order to detect Parkinson’s disease at an early stage, numerous indicators were used in this study, including Rapid Eye Movement and olfactory loss, cerebrospinal fluid data, and dopaminergic imaging markers. A comparison of the proposed deep learning model with twelve machine learning and ensemble learning methods based on relatively small data sets of 183 healthy individuals and 401 early Parkinson’s disease patients reveals that the designed model achieves the highest accuracy, 96.45 percent on average. We present the feature importance of the PD detection procedure based on the Boosting approach in addition to identifying the PD. Traditional diagnostic procedures, on the other hand, may suffer from subjectivity because they rely on the interpretation of motions that are sometimes subtle to human sight and hence difficult to define, potentially leading to misdiagnosis. Meanwhile, early non-motor symptoms of Parkinson’s disease may be moderate and caused by a variety of other illnesses. As a result, these symptoms are frequently missed, making early PD diagnosis difficult. To overcome these challenges and improve PD diagnosis and assessment procedures, machine learning algorithms for the viii classification of PD and healthy controls or patients with similar clinical presentations have been used. A total of 209 papers were included, with pertinent material retrieved and provided in this review, with an examination of their aims, data sources, data kinds, machine learning methods, and associated outcomes. These results show that machine learning methodologies and novel biomarkers have a strong potential for adaption in clinical decision-making, leading to more systematic, informed PD diagnosis.
  • Thumbnail Image
    Item
    Open Access
    Sleep Apnea Detection From Raw Ecg Signal Using Deep Learning And Machine Learning
    (North South University, 2023) Salem Shamsul Alam; Sumit Saha; Md. Shahriar Hussain; 1931849642; 1931415042
    Sleep apnea, a prevalent yet underdiagnosed sleep disorder, necessitates robust and accurate diagnostic tools. In this project, we undertook an in-depth exploration of machine learning (ML) and deep learning (DL) models for sleep apnea detection, specifically utilizing raw electrocardiogram (ECG) signals. Our comparative analysis encompassed a range of ML models, including Random Forest, Logistic Regression, Decision Tree, AdaBoost, and XGBoost, and a specialized 1D-CNN model within the DL domain. Results underscore the exceptional performance of the 1D-CNN model, achieving a remarkable accuracy of 99.56%, sensitivity of 96.05%, and specificity of 99.66%. This outperforms traditional ML models, signifying the prowess of DL in extracting intricate patterns from raw ECG signals for accurate sleep apnea detection. The 1D-CNN model's ability to discern subtle features proves crucial for accurately identifying apnea events. Our study not only emphasizes the effectiveness of the 1D-CNN model for sleep apnea detection and highlights the transformative potential of deep learning in healthcare diagnostics. This research contributes valuable insights into the optimal choice of models for sleep apnea detection, paving the way for enhanced diagnostic accuracy and improved patient care.
  • Thumbnail Image
    Item
    Open Access
    Surface Electromyography (sEMG) Based Cost-Effective Prosthetic Arm
    (North South University, 2022-01-01) RIDHWAN HOSSAIN AFRIDI; MD. ALIM ULLAH CHOWDHURY; SABINA RAHMAN; Md. Shahriar Hussain; 1822074043; 1811412043; 1822065043
    Surface Electromyography (sEMG) signals are biomedical signals that represent electrical currents generated during muscle activity, and our central nervous system controls these signals. As a result, we can develop a prosthetic arm using EMG technology. Currently, there are many myoelectric prosthetic arms available commercially. However, they are costly for developing countries like Bangladesh, India, and Pakistan. So we developed a cost-effective circuit to extract the electromyography signal from the skin. The signal from this detector circuit is imported in MATLAB using Arduino Uno R3 for signal analysis, such as Fast Fourier Transform (FFT) and Wavelet Transform (1-D), for the classification of features from the signal. We successfully detected two distinct features during the analysis: the Rest position and the Fist position of the hand. So we used the ESP32 microcontroller for the practical implementation of the system as it has better ADC (12-Bit), higher clock speed (80 MHz), better PWM, and lower power consumption. We used a threshold level to distinguish between the Fist and Rest positions. The threshold value was determined using the trial and error method, and this value may vary from person to person. Finally, the ESP32 drives five servo motors fitted inside the InMoov open-source 3D arm after determining the hand position
  • Thumbnail Image
    Item
    Embargo
    Web application for monkeypox disease detection using deep learning
    (North South University, 2022) AHMAD SAMIN SHADMAN; SUMAIYA SHARMEEN; Md. Shahriar Hussain; 1811437042; 1731500042
    The monkeypox virus might become the next big pandemic, like the COVID-19 pandemic, if it is not monitored and controlled correctly. Monkeypox has some similarities to measles and chickenpox, making it very hard to test for it and give a diagnosis in the early stages. A polymerase chain reaction (PCR) test must be used to test for monkeypox properly. This study aims to detect monkeypox accurately using some popular deep-learning models and then compare the results. We used the “Monkeypox Skin Lesion Dataset (MSLD).” Data augmentation has been done to the data to increase the number of images. A web-based prototype application is to be developed where an image can be uploaded, and a prediction will be given if the image is either monkeypox or not. The model used in the web application is the VGG-16 model which identifies monkeypox images with an accuracy of 99%.