1 Oct 2020

Evaluation of Unmanned Aircraft System Coastal Data Collection and Horizontal Accuracy: A Case Study at Garden City Beach, South Carolina

Abstract: The US Army Corps of Engineers (USACE) aims to evaluate unmanned aircraft system (UAS) technology to support flood risk management applications, examining data collection and processing methods and exploring potential for coastal capabilities. Foundational evaluation of the technology is critical for understanding data application and determining best practices for data collection and processing. This study demonstrated UAS Multispectral (MS) and Red Green Blue (RGB) image efficacy for coastal monitoring using Garden City Beach, South Carolina, as a case study. Relative impacts to horizontal accuracy were evaluated under varying field scenarios (flying altitude, viewing angle, and use of onboard Real-Time Kinematic–Global Positioning System), level of commercial off-the-shelf software processing precision (default optimal versus high or low levels) and processing time, and number of ground control points applied during postprocessing (default number versus additional points). Many data sets met the minimum horizontal accuracy requirements designated by USACE Engineering Manual 2015. Data collection and processing methods highlight procedures resulting in high resolution UAS MS and RGB imagery that meets a variety of USACE project monitoring needs for site plans, beach renourishment and hurricane protection projects, project conditions, planning and feasibility studies, floodplain mapping, water quality analysis, flood control studies, emergency management, and ecosystem restoration.

23 Apr 2020

PUBLICATION NOTICE: Use of Convolutional Neural Networks for Semantic Image Segmentation Across Different Computing Systems

ABSTRACT: The advent of powerful computing platforms coupled with deep learning architectures have resulted in novel approaches to tackle many traditional computer vision problems in order to automate the interpretation of large and complex geospatial data. Such tasks are particularly important as data are widely available and UAS are increasingly being used. This document presents a workflow that leverages the use of CNNs and GPUs to automate pixel-wise segmentation of UAS imagery for faster image processing. GPU-based computing and parallelization is explored on multi-core GPUs to reduce development time, mitigate the need for extensive model training, and facilitate exploitation of mission critical information. VGG-16 model training times are compared among different systems (single, virtual, multi-GPUs) to investigate each platform’s capabilities. CNN results show a precision accuracy of 88% when applied to ground truth data. Coupling the VGG-16 model with GPU-accelerated processing and parallelizing across multiple GPUs decreases model training time while preserving accuracy. This signifies that GPU memory and cores available within a system are critical components in terms of preprocessing and processing speed. This workflow can be leveraged for future segmentation efforts, serve as a baseline to benchmark future CNN, and efficiently support critical image processing tasks for the Military.