Now showing items 1-6 of 6

  • Converting 2D motion into 3D world coordinates in the case of soccer players video.

    T.E.I. of Crete, School of Engineering (STEF), PPS in Informatics and Multimedia
    Authors: Charalapaki, Eirini
    Thesis advisor: Malamos, Athanasios
    Publication Date: 2018-07-17
    One of many important tasks in computer vision applications is real time object detection and tracking. Motion information and visual features, such as shape, color and texture, can be used to extract information from a ...
  • Development of a control system for a 2 DOF robotic arm which includes machine vision.

    T.E.I. of Crete, School of Engineering (STEF), IPPS Advanced Manufacturing Systems, Automation and Robotics
    Authors: Froudarakis, Konstantinos
    Thesis advisor: Kavvousanos, Emmanouil
    Publication Date: 2016-07-15
    The purpose of this paper is to develop a Machine Vision system, with only one camera, which can identify an object on a ballistic trajectory, calculate the point of impact on the plane of the robotic arm, and intercept ...
  • Object recognition using depth maps and machine learning technologies.

    T.E.I. of Crete, School of Engineering (STEF), Department of Informatics Engineering
    Authors: Kounalakis, Tsampikos
    Thesis advisor: Triantafyllidis, Georgios
    Publication Date: 14-04-2011
    This paper presents a novel system that is fusing efficient and state-of-the-art techniques of stereo vision and machine learning, aiming at object detection and recognition. To this goal, the system initially creates depth ...
  • Anticipating human activities for reactive robotic responses.

    T.E.I. of Crete, School of Engineering (STEF), PPS in Informatics and Multimedia
    Authors: Andrianis, Athanasios
    Thesis advisor: Papadourakis, Georgios
    Publication Date: 2016-07-14
    In this Master thesis the objective was to implement an activity recognition application in order to have the first step for activity anticipation, using MATLAB. The CAD-60 dataset. We implement four different ways with ...
  • Iris detection and applications.

    T.E.I. of Crete, School of Engineering (STEF), Department of Informatics Engineering
    Authors: Zouzoulas, Charalampos; Theodorou, Vasilios
    Thesis advisor: Triantafyllidis, Georgios
    Publication Date: 25-05-2011
    In this final year project, we describe ways of detecting the iris with the use of computer vision techniques. Then the development of an application is presented, where the cursor movement is achieved accordingly to the ...
  • Deep learning framework, for image classification applications.

    T.E.I. of Crete, School of Engineering (STEF), PPS in Informatics and Multimedia
    Authors: Frangiadoulis, Alexandros
    Thesis advisor: Vidakis, Nikolaos
    Publication Date: 2016-08-23
    In this master thesis, we design and construct a Deep-Learning Framework, which is a model for building image classification applications. We demonstrate the “CNNs Tester” and a web-application structure. “CNNs Tester” ...