Thesis (Bachelor of Science in Computer Science)

Browse

Author: search.filters.author.Sakib, SadmanHas files: Yes

Search Results

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    A modern technique to detect potholes by Computer Vision and Deep Learning
    (BRAC University, 2022-09) Saif, Muntasir Mahmud; Badsha, Tanvir; Khan, Mohammed Arman; Sakib, Sadman; Bin Akbar, Rafeed; Karim, Dewan Ziaul
    Roads are connecting lines between different places and are used in our daily life but anomalies in road surface not only impact road quality but also affect driver safety, mechanic structure of the vehicles, and fuel consumption. Several approaches have been proposed to automatic monitoring of the road surface condition in order to assess road roughness and to detect potholes. Potholes are one of the main reasons behind the occurrence of road accidents. According to a report submitted by The Roads and Highways Department (RHD), around 25% roads of Bangladesh under the RHD across the country are in ”poor, bad or very bad” condition. This causes a lot of hassle and issues on the road for both humans and vehicles. Very often be cause of these potholes road accidents occur. Techniques for detecting potholes on road surfaces are being developed to provide real-time or offline vehicle control (for driver assistance or autonomous driving) as well as offline data collecting for road repair. For these reasons, researchers have looked into ways for detecting potholes on roads all over the world. This paper begins with a quick overview of the area before categorizing developed strategies into various groups. Then, by developing method ologies for automatic pothole detection, we present our contributions to the field. For this reason, we propose a deep learning approach that allows us to automatically identify the different kinds of road surface and to automatically distinguish potholes from destabilizations produced by speed bumps or driver actions. The system can detect potholes in different environments, lighting and weather conditions. We have trained and tested our model with a custom dataset which contains raw 3000 images with 1500 normal road images and 1500 images with potholes using deep learning algorithms. We have augmented these images and turned them into 120000 images so that the model can understand any image input in any scenario. In particular, we have analyzed and applied different deep learning models such as convolutional neural networks (CNN) and Yolov4. With these models we have achieved 97.35% accuracy with the CNN model and 87.6% accuracy with the YOLOv4 model.
  • Thumbnail Image
    Item
    A convolutional neural network based model with improved activation function and optimizer for effective intrusion detection and classification
    (BRAC University, 2020-04) Kabir, Solaiman; Sakib, Sadman; Hossain, Md. Akib; Islam, Safi; Hossain, Muhammad Iqbal
    In today's world, technological advancements have entangled our nancial, social and many more other aspects of lives to the internet or some network. Moreover, with the development of IoT technologies, it has spread over to our transportation, home-appliances and more devices. It is also a security risk because all of our sensitive and private knowledge on the Internet is exposed to a growing amount of cyber-attacks. An Intrusion Detection System can identify a cyber-attack while it is ongoing or prior to it. We are conscious of the evolving Machine Learning and Deep Learning developments, the most sophisticated multi-functional methods created by humans that can be utilized to overcome this issue. Alongside identi fication, precise classi cation of intrusion is of considerable signi ficance for the administrator to take decisive actions. In this study, we have used the dataset CIC-IDS-2018 that is the biggest and most recent labeled dataset of intrusions. This dataset comprises of six varieties of attacks. Our thesis proposes a CNN Model with mish activation function and Ranger optimizer. The model reaches an accuracy of 0.989 that is the highest in multiclass classification with this dataset.