Browsing by Author "Dr. Atiqur Rahman"

Now showing 1 - 4 of 4
  • Thumbnail Image
    Item
    Open Access
    A Comparative Study of Machine Learning Techniques for Autism Spectrum Disorder (ASD) Detection
    (North South University, 2022) Zahid Hasan; Towsif Muhtadi Khan; Dr. Atiqur Rahman; 1911509042; 1911576042
    Autism Spectrum Disorder (ASD) is a developmental impairment caused by brain differences. ASD patients may have a recognized difference, such as a genetic disease [1]. Autism can be diagnosed at any age and is referred to as a "behavioral disorder" since symptoms often develop within the first two years of life [1]. In this study, we have implemented different classifier algorithms such as Decision Tree Classifier, Logistic Regression, and Multi-layer Perceptron classifier (MLP Classifier) for detecting ASD in three types of people - adults, adolescents, and children. The proposed techniques are evaluated on publicly available three different non-clinically ASD datasets [2]. After applying various machine learning techniques and handling missing values, results strongly suggest that ANN-based prediction models work better on all these datasets with higher accuracy of 98.58%, 98.30%, and 95.24% for Autistic Spectrum Disorder Screening Data for Adults, Children, and Adolescents, respectively.
  • Thumbnail Image
    Item
    Embargo
    AI Dekho (Smar t spect acle for visually challenged people)
    (North South University, 2023) Shajeda Par vin; MD Shadman Zar if; Dr. Atiqur Rahman; 1931313042; 1931796042
    Our study addresses t he hardships experienced by individuals lacking good vision. The cur rent technology t hat helps t he visually impaired people is quite expensive and does not work well for recognizing people’s faces and objects around t hem. The paper introduces a new idea of g lasses t hat can help visually impaired people. As t he wear able market expands, t he primary aim is to enhance people’s lives through t he use of wear able technology. The primary goal of t his project is to provide assistance to visually impaired individuals through t he utilization of g lasses equipped wit h a distinctive camera. The camera helps t hem to identify people and objects, which is important for t heir daily lives. A collection of pictures showing regular people is put together to help wit h identifying t hem. This met hod uses cameras to help people who can’t see well. The main goal is to make special g lasses t hat can recognize faces and objects in real-time using library called OpenCV. The system has two parts, one t hat faces t he user and one t hat connects to a computer. The findings indicate its ability to identify individuals and surroundings, although concerns remain regarding its speed and data transmission methodologies. It aims to redevelop t he wear able wit h more features and cloud computing technologies.
  • Thumbnail Image
    Item
    Open Access
    Detection of Violent Activities Using Deep Learning Algorithms
    (North South University, 2022) Tasmiah Sarker; Fayeeka Simran; Zahiduzzaman Anik; Dr. Atiqur Rahman; 1912844642; 1911656642; 1632091042
    The creation of a method for violence detection in surveillance footage using automatic analysis is crucial. In this study, we propose a deep neural network to recognize violent videos. A convolutional neural network and an ImageNet model that has already been trained are used to extract frame level characteristics from a movie. Then, using a long short-term memory variation that makes use of fully connected layers and leaky rectified linear units, the frame level features are aggregated. Convolutional neural networks are capable of recording localized spatio-temporal information that allow the analysis of local motion in the video, in addition to long short-term memory. On three common benchmark datasets, the accuracy of recognition is used to further assess the performance. We also contrasted the findings of our system with those from other methodologies to ascertain the capabilities of our proposed model. The suggested solution outperforms cutting-edge techniques while processing the videos in real-time.
  • Thumbnail Image
    Item
    Open Access
    Smart Traffic Control System
    (North-south University, 2022-01-20) Syed Asim Anwar; Md. Hasnat Alam; Fatima Tuz Zohura; Dr. Atiqur Rahman; 1711850642; 1712610642; 1711335642
    This paper presents the implementation of smart traffic control systems using computer vision. In present days with the expansion of innovations, specialists are always looking for innovatives systems for traffic control. Nowadays traffic jams have become one of the major issues in Bangladesh. In our country, the vehicle density on the road is slowly becoming greater and much more uncomfortable. Road congestion has a major effect on the normal journey of people and restricts the economic development of society. And the main reasons for this type of road congestion are due to poor traffic management. In this paper, we propose a smart traffic control system by measuring the traffic density of the road by real time detection and image processing. So, we created a vehicle detection model to count the number of vehicles approaching a signal and we used a proper algorithm to measure the signal time based on the number of vehicles on the road.