Browsing by Author "Islam, Sadman Saumik"

Now showing 1 - 4 of 4

Autonomous Virtual Vehicle Using Reinforcement Learning
(Daffodil International University, 2019-12-05) Islam, Sadman Saumik; Alam, Samia Binta
Artificial intelligence plays a vital role in self driving vehicle. The direction in which the automotive industry is headed it is expected that autonomous vehicles capable of driving without human supervision will be released to market in the next decade. The development of such intelligence is fairly in its early stages. The problem of creating such intelligence is that the real-world environment is ever-changing. To solve this problem the autonomous vehicle needs to have such intelligence that it can cope with the changing real-world environment. We though can achieve such intelligence by simulating an autonomous self-driving agent in a virtual 2D platform where it will be able to follow the track on its own. Using supervised learning to solve this problem will not be an efficient approach to this problem because no matter how much training and testing is done, it will not be able to keep up with the dynamic real-world environments. Therefore, we are proposing a novel approach for creating autonomous AI powered virtual vehicle using Reinforcement Learning.
eExpense
(5th International Conference on Electrical Engineering and Information and Communication Technology, IEEE, 2019) Sabab, Shahed A.; Islam, Sadman Saumik; Hossain, Monir
Tracking regular expense is a key factor to maintain a budget. People often track expense using pen and paper method or take notes in a mobile phone or a computer. These processes of storing expense require further computations and processing for these data to be used as a track able record. In this work, we are proposing an automated system named as eExpense to store and calculate these data. eExpnese is an application that runs on Android smartphones. By using this application, users can save their expense by simply scanning the bills or receipt copies. This application extracts the textual information from the receipts .and saves the amount and description for further processing. It also monitors user’s income by tracking the received SMS’s from the user’s saving accounts. By calculating income and expense it produces the user’s balance in monthly and yearly basis. Overall, this is a smart automated solution for tracking expense.
Estimating Energy Expenditure of Push-Up Exercise in Real Time Using Machine Learning
(Springer, 2023-06-11) Uddin, Md. Shoreef; Islam, Sadman Saumik; Hussain, M. M. Musharaf
The Covid-19 pandemic has nearly brought the globe to a standstill. However, we were able to adjust to the circumstance with the aid of computer technology by working remotely from home. Health and fitness have grown to be top priorities during these tumultuous times when people are confined to their homes. By completing certain easy physical activities that don’t require any special equipment and can be done at home, a person can maintain good health and fitness. Furthermore, the detection and recognition of human body motions or gestures is not a new concept when using artificial intelligence. Analysis of human body movement is now quicker and easier because of the development of real-time detection and identification technologies like YOLO. In this study, we have employed YOLO V4 as an AI helper to detect and identify push-ups from a real-time video stream recorded from a webcam or a smartphone camera that may be used to aid with push-ups. In order to keep the system affordable, we are recommending an approach that can identify pushups and estimate energy usage in real-time without the use of additional sensors or other wearable gadgets."
Hand Swiftera
(4th International Conference on Electrical Engineering and Information and Communication Technology, 2019) Sabab, Shahed Anzarus; Islam, Sadman Saumik; Hossain, Mainul; Shahreen, Mahbuba
In the last decade, the trends in Human-Computer Interaction have changed dramatically with the invention of new and emerging technologies and devices. Controlling computer without the traditional peripherals is one of that trends. Our system is a new step to reduce the use of conventional mouse and keyboard. The proposed approach is a real-time hand tracking process based on webcam to navigate a computer. With the help of Kalman filter mouse pointer flickering was reduced and gained a better accuracy and smoothness for mouse pointer movement. The system focuses on navigating computers, reading documents and controlling multimedia like videos, pictures etc. So, we designed our system for three different modes as, Navigation, Reading, and Multimedia mode. We implemented Leap Motion device for switching to different modes and developed different types of hand gestures as swiping and tilting hand. This approach can ensure user comforts and prevent ergonomic injuries. The developed system Hand Swifter - not only improves better computer using experience but also helps to prevent repetitive strain injuries caused by traditional input devices such as mouse and keyboard.