Night vision with unmanned surface vehicle (USV)

Abstract

Night vision technologies are implemented in Unmanned Surface Vehicles (USV) mainly for detection, recognition identification. Night vision technologies are Image Intensification, Active Illumination and Thermal Imaging. In this Final Year Project (FYP), this report focuses on my research on types of cameras that accept low lighting conditions. Besides that, this research also involves sourcing code on a deep learning method for transferring night-time images to daytime images. Besides that, this research also includes object detection algorithms. The source code for transferring night-time to daytime images uses ToDayGan, a deep learning method. The types of object detection algorithms are Histogram of oriented gradients(HOG), Convolutional Neural Networks (CNNs) and You Only Look Once (YOLO).