2021 International Conference on Nascent Technologies in Engineering (ICNTE 2021) 978-1-7281-9061-7/21/$31.00 ©2021 IEEE Automated Surveying for Construction Engineering 1 Mohd. Jamaluddin Khan Department of Information Technology Fr. C. Rodrigues Institute of Technology, Vashi Navi Mumbai, India jkhan266@gmail.com 3 Lakshmi Gadhikar Department of Information Technology Fr. C. Rodrigues Institute of Technology, Vashi Navi Mumbai, India lakshmi.gadhikar@fcrit.ac.in 2 Jim Caleb R. Department of Information Technology Fr. C. Rodrigues Institute of Technology, Vashi Navi Mumbai, India jimcaleb57@gmail.com 4t Sharon Mehershahi Department of Information Technology Fr. C. Rodrigues Institute of Technology, Vashi Navi Mumbai, India sharon99mary@gmail.com Abstract — India being a land of diverse landscapes and one of the largest geographical nations inevitably faces the need to develop roads for easing transportation, facilitating human resource services and to bring the entire country on a common economical platform. However, construction of roads in rough terrains and remote areas still remains a challenge and involves a lot of time. Conventional Highway Engineering based surveying and road layout designing involves a lot of manpower as well. Here, the solution of automated surveying helps out. Automating the rationales of landscape surveying using UAV (Unmanned Aerial Vehicles) speeds up the planning, surveying process and at the same time eases the generation of EIA (Environmental Impact Assessment) Report. By capturing the real time images of the target site and then processing them on a remote server to generate a VARI (Variable Atmospheric Refractive Index) value which is a type of vegetation index along with water body and settlement segmentation, a summary report gets automatically generated in minutes. This helps to cover some of the crucial factors associated with EIA reports in no time. Also, by using the location’s height-map image, a proper terrain in the form of DEM (Digitally Elevated Model) to plan the surveying is generated which also helps in understanding the site’s topology better visually. Keywords — Highway Engineering, Automated Surveying, Variable Atmospheric Refractive Index (VARI), Environmental Impact Assessment (EIA), Unmanned Aerial Vehicle (UAV), Digitally Elevated Model(DEM) I. INTRODUCTION India has the second-largest road network in the world. According to 2009 estimates by Goldman Sachs, India will need to invest US$1.7 trillion on infrastructure projects before 2020 to meet its economic needs, a part of which would be in upgrading India's road network [1]. Roads play a very important role in the transportation of goods and passengers for short and medium distances. Thus, with roads being an integral part of economy and development, improvising the methods and rationales of highway engineering will in turn prove beneficial for a rapid industrialization and consequent modernization. It will help in providing an impetus to the GDP and ensure an all-inclusive growth. With the motive to bring the villages at par with developed cities and provide them modern day facilities, the villages need to be accessible through land routes as early as possible. The transportation projects need to be completed at a faster rate for a stable economy, resistant to global financial crisis. Surveying, planning and estimation all in itself takes a significant duration of time. The solution presented here is an automated unmanned aerial vehicle (UAV) or drone for performing the surveying operation. The drone is deployed to conduct the aerial survey of the region and then perform the analysis of extent, also understand the vegetation status, water-bodies and settlements present in the region of study. It thus performs the environmental impact assessment. Beforehand, using the open source height-map images we get a Digitally Elevated Model generated which helps to understand the topology and terrain of the site better. This can enable the engineering team to directly proceed towards the layout generation and foundation laying process in no time. II. RELATED WORK Generating Digitally Elevated Models (DEM) using Python Photogrammetry Toolbox [2] is one of the few open- source techniques for generating the terrain of the site. The solution uses a number of high performance data-science Python libraries. The toolbox is used for 3-D reconstruction purposes of models. The technique involves pre-processing of close range images captured using drones followed by feeding the images to the PPT and immediate post-processing to obtain suitable terrain. Their test results prove that the open source toolbox proves to be highly reliable and cost-effective. However, Python Photogrammetry Toolbox is deprecated and is no longer available as a standalone application. At the same time, many modern day photogrammetry software programmes are available which have python plugins. Some of these are Meshroom, Visual SFM, ColMap, etc. [3]. Also, there are applications, which can render heightmap data images obtained from sources like Open Street Maps or Google Maps into 3-D terrain. Using this approach of Open Street Maps based heightmap images and their rendering, we will be able to reduce the resource consumption involved in generating the DEM of a region. The approach used to detect water bodies from the Landsat images using the Perceptron model [4] takes a set of feature vectors as input which are nothing but pixel data generated by water-bodies. The output is nothing but a binary value classifying the water-body region from the non-water-body region. The weighted sum acquired for each pixel is compared with the threshold value and in this way the classification is 2021 4th Biennial International Conference on Nascent Technologies in Engineering (ICNTE) | 978-1-7281-9061-7/21/$31.00 ©2021 IEEE | DOI: 10.1109/ICNTE51185.2021.9487670 Authorized licensed use limited to: Bannari Amman Institute Of Technology - Erode. Downloaded on September 01,2021 at 23:56:55 UTC from IEEE Xplore. Restrictions apply.