Browsing by Author "Riasat Khan"

Now showing 1 - 6 of 6
  • Thumbnail Image
    Item
    Open Access
    A Medical Community Android App, Detect COVID 19 and Pneumonia Using Deep-learning
    (North South University, 2021) Shuva Chowdhury; Istiak Ahamed Saif; Faijul Abedin; Amirul Ahsan Simon; Riasat Khan; 1731245042; 1712196642; 1711316642; 1632129642
    COVID-19 is the biggest headache for the whole world, including detecting COVID-19-affected patients. Early detection of COVID-19 may aid in the development of a treatment strategy and disease containment decisions. Also, a community through application among doctors, nurses, and patients can reduce deprivation of treatment and health care services. In this paper, we make a medical community Android application for doctors, nurses, and patients that can detect COVID-19 from chest X-ray photographs developed using a convolutional neural network deep learning algorithm (VGG16). The COVID-19, Pneumonia, and standard chest X-ray images are collected and joined from a public source, Kaggle. 9000 chest X-ray photographs were used for training, including 3000 COVID-19 chest X-ray photographs, 3000 Pneumonia chest X-ray photographs, and 3000 standard chest X-ray photographs. For testing, 3000 chest X-ray photographs were collected, with 1000 COVID-19 chest X-rays, 1000 Pneumonia chest X-rays, and 1000 normal chest X-rays. The accuracy of our training is 98 %, while the accuracy of our validation is 95%.
  • Thumbnail Image
    Item
    Open Access
    Automated Parking Entrance System Using Different Approaches
    (North South University, 2020) Md.Naimul Islam Suvon; Mehebuba Ferdous; Riasat Khan
    Automatic license plate identification in today's world plays a vital role in vehicle tracking and organization. Our proposed model of automation in the detection and recognizing vehicles through the use of number plate computerization is expected to create a new scope of evolution for large cities. The machine can be used for the parking system of motor vehicles, as well as to collect tolls. The detection of the Bangla license plates from different cities and multi-class vehicles is the first step of the proposed system. The license plate detection has been performed with mathematical morphological approach and the Faster R-CNN architecture. Next, the Tesseract Optical Character Recognition system, in conjunction with the Bangla character recognition model, has been used for vehicle indexing. We are also planning to make an application for the parking lot system. That will manage all the information about the parking lot system. Numerical results demonstrate that the accuracies of license plate detection for the computer vision and Faster R-CNN are 95% with limitation and 98% without any limitation, respectively. Finally, our model achieved an accuracy of 96% for character recognition with some limitation.
  • Thumbnail Image
    Item
    Open Access
    Determining the Radio Link Losses in different medium and Field Measurement in the Context of Bangladesh
    (North South University, 2021) Mohammad Akib Javed Mojumder; Sheikh Shah Newaz; Md. Nihal Al Rafi; Riasat Khan; 1531175043; 1632206043; 1721703043
    This paper contains estimated information about the different purpose behind radio link path loss and why it is essential to identify this explanation and how radio link path loss influence the systems, the best approach to tackles those issues. Radio link path loss happens due to various when an item shows up in the way. As needs are, radio path loss figuring is utilized in many radio and wireless devices for deciding signal quality in different areas. These remote review devices are, as a rule, progressively helps to figure out how signal qualities will be, before introducing the hardware. In spite of the fact that expectation can be genuinely exact for the free space condition, for genuine earthly applications, it seems difficult as there were numerous problems to consider, and it isn't generally conceivable to gain a precise rating of the impacts that we will have. This analyzing help to design better radio transmission system with less distortion and loss. Hence the accuracy of information will not be flawless because of the complexity of the obstacles. Here is some field measurement in the context of Bangladesh radio link. Our result shows the way of improvement if we avoid the various factor that cause radio link path loss.
  • Thumbnail Image
    Item
    Open Access
    Photo-To-Cartoon Translation with Generative Adversarial Network
    (North South University, 2022) Istiaque Ahmed; Kazi Md. Ifthekhar Uddin; Rakibul Hasan; Riasat Khan; 1812420042; 1811019042; 1811194042
    Cartoons are a popular art form in our daily lives, and the ability to automatically create cartoon graphics from photos is highly desired. Cartoon images have a more vibrant and lively appearance than traditional naive pictures. This study aims to explain the process of translating real-world photos into cartoon-like images. While converting pictures to cartoons, there were a few difficulties, including fine hair edges, mismatched colors, and texture concerns. Photos were converted to cartoon-style images using generative adversarial networks (GAN). Various neural network-based GAN networks, DCGAN, CycleGAN, and AnimeGAN, have been applied in this work for cartoon conversion. Among them, CycleGAN performs better in transforming actual photographs into colorful, eye-catching cartoons. This project's approach is based on learning-based methodologies, which have lately gained popularity for stylizing images in artistic forms like painting. The results may be used to convert real-world photographs to high-quality cartoon graphics quickly. This project provides a web API that contains training weights derived from the models outlined below. Based on that API, we created a web app that converts real-world images into high-quality cartoon graphics for various cartoon styles. In these experiments, it outperforms state-of-the-art approaches to producing high-quality cartoon graphics from real-world photos. Numerical results show that the CycleGAN approach has the lowest training time per epoch and requires the minimum number of trainable parameters.
  • Thumbnail Image
    Item
    Open Access
    Speaking System for Mute And Deaf People
    (North South University, 2021) Zunaid Zaohar; Asmany Akter Sornaly; Atik Shahariar Patwary; Riasat Khan; 1531342042; 1621179043; 1512384042
    The major goal of this paper is to provide social service and to implement an automatic speaking system for the deaf and dumb. Communication is one of the most important aspects of human life. Every living being communicates in their own unique way. What kind of existence would we have if we couldn't speak or listen? So, in general, we can observe that many people in our community are not as fortunate as we are. There are certain individuals who are unable to speak or listen. As a result, the primary goal of our initiative is to assist people who are unable to speak or listen. Because many of us do not understand sign language, it will make it easier for people with disabilities to communicate with us and share their experiences. It will have a significant impact on healthcare and education. We're utilizing wearable gloves so that anyone can do it. An intelligent sign language recognition system for deaf individuals was developed in this study. Four flex sensors, an accelerometer, a speaker module, gloves, an Arduino Uno, an LCD screen, a GSM 800L module, and a Lithium Rechargeable Battery were utilized in this project. Our hardware tools after connecting to basic C programming runs successfully.
  • Thumbnail Image
    Item
    Open Access
    Voice Controlled Color Sorting Robot for Picking-Up and Placing Objects
    (North South University, 2020) Arka Datta; Mahmudur Rahman; Kaniz Fatema; Riasat Khan; 1521423043; 1520386043; 1520064045
    Humanity has always attempted to give life to its artefacts as features in an attempt to find substitutes for itself to fulfil its directions and also to function in a hostile setting. A machine that looks like a human arm and works like it is the popular concept of a mechanical arm. The industry is moving from the current state of automation to robotization. Increase productivity and consistent quality production. The industrial robots of today may not look at all like a human being. However, the whole research aims to provide more and more anthropomorphic, and human-like features. Managing certain specific responsibilities, like sending a robotic car to hazardous settings for chemical analysis specimens. A typical robotic car can travel and cross obstacles across distinct terrains. A robotic manipulator or just a mechanical arm is one robot type that is commonly used in the industry. It is an open or closed kinematic chain of stiff links interconnected by mobile joints. In some environments, links can be considered as waste to match human anatomy. Upper arm with joint and forearm of the shoulder and elbow. At the end of the arm, a wrist joint connects an end-effector that can be an tool with which to function, or a gripper or any other device. The primary objective of this project is to create an industry/restaurant pick-and-place robot as an option for human employees. Compared to others, this scheme works correctly. The one portion of our project is the robotic vehicle and the other portion is the robotic arm component. The robotic vehicle is governed by voice, while the other is controlled by a joystick, which is the robotic arm. Thus it makes the pick and place robotic arm perfectly mobile. This project results in our hard work and commitment.