Remote Sensing Technologies and Applications in Urban Environments VII

We present the concept of a mobile measurement platform paired with an end-to-end data processing chain that enables analysis of multimodal sensor data in real time for smart city mapping. The proposed system can be integrated on mobile platforms into the everyday traffic in urban environments. The online pre-processed and compressed information can then be used to directly update a cloud-based digital twin. This enables the creation of a virtual image of entire cities and generates data that can be used for real time Urban Information Modeling, and thus a valuable planning tool to provide up-to-date information at any time. The generated data form the basis for decision-making on improving mobility flows for smart transportation systems and autonomous vehicles and the survey of infrastructure and vegetation for a sustainable urban development. The proposed concept is achieved using energy efficient embedded sensors and processing units in combination with computational optimized software architectures close to the sensors.

The automated segmentation of photogrammetric or LiDAR point cloud data offers the basis for further post-processing. However, the quality of the segmentation highly depends on the amount and quality of annotated training data and the applied segmentation method. This paper gives an overview and compares state-of-the-art point cloud segmentation methods including both 2D, 3D and combined approaches. Furthermore, current annotation tools and methods are evaluated regarding the accuracy, the number of classes, and time effort. Moreover, current strategies and methods to extend training data with augmented or synthetic data are listed and assessed. Besides, we present a concept for a specific segmentation task based on a trade-off of the previously listed methods. The task includes the segmentation and partial annotation of colourized point cloud data of two urban areas in Freiburg Germany, acquired from a driven and flying platform. Lastly, we present the required steps to extend the segmentation task from one city section to an overall city with diverse areas and to adapt to different sensor systems leading to different point cloud specifications. We conclude this paper by identifying challenges and required research in the field of point cloud segmentation and annotation.

An effective maintenance plan of railways bridges needs suitable, user-friendly tools for preventive inspection that otherwise requires a large amount of resources and it is time consuming. In this paper we present a series of experiments conducted with the aim of developing an ICT tool for supporting the inspection of railways bridges. The developed tool integrates a full range of diverse data and guarantees easy access for their storage and consultation. The data integrated in the ICT platform includes images acquired with diverse techniques (3D scanning, high definition photography, photogrammetry, thermography, fluorescence LIDAR) and data from traditional survey methods.

A modern integrated transport system is a central point to ensure a well-functioning Infrastructure. Therefore, it is not only important to plan new parts of the infrastructure with modern technology but also to digitalize existing infrastructure elements. Bridges in particular are an important part of our infrastructure but monitoring them is still in its infancy. For now, all information of bridges such as planning documents, statics, status reports of bridge examination, etc. are collected in decentral locations of the owner or operator. The existing information is available in a wide variety of formats e.g. pdf files, plans on paper, scanned paper plans, digitally created plans, SAP-data and photos. We tackled this problem of non-uniform and decentralized data management within the research project mdfBIM. We analyzed the process of creating digital models of railway bridges regarding automation potential. In the developed process chain, various data streams are merged. LiDAR scans, photogrammetric pointclouds and existing CAD plans are homogenized and merged into a complete BIM model. Here we present the results based on a railroad bridge in Hanover and provide an insight into the necessary processes and lessons learned.

This paper aims to analyze the evolution of land use patterns in the main European metropolitan areas. For this, the main metropolitan areas with more than 1 million inhabitants will be delimited using nighttime remote sensed images. In authors previous works, it had been delimited urban (compact and sprawled), rurban and rural landscapes around the world. There are 404 metropolitan areas with more than one million inhabitants and 60 are in Europe. In this sense, the evolution of land covers between 2000 and 2018 of the main European metropolises will be studied: The Belgian/Dutch agglomeration, the diffuse landscape of Northern Italy, London, Paris, the Rhine-Ruhr, the English central agglomeration, and the metropolitan areas of Madrid, Barcelona, Rome and Naples. The evolution of land occupation will be studied, determining an "environmental quality index", which will allow knowing the process of improvement or deterioration of environmental quality.

Mobile mapping vehicles are rapidly gaining in importance as they accelerate the digitization of road surroundings as required by municipalities for infrastructure projects. The evolution of laser scanners and high-resolution cameras demands the development of new strategies for the handling of large amounts of data. In this paper, a measurement vehicle is presented that handles up to 4 GB/s of raw data at driving speeds up to 144 km/h, yielding 3D point clouds and high-resolution panoramic images. A sophisticated on-board data processing pipeline was developed that manages and stores the data.

Unmanned Aerial Vehicles (UAV) provide increased access to unique types of urban imagery traditionally not available. Advanced machine learning and computer vision techniques when applied to UAV RGB image data can be used for automated extraction of building asset information and if applied to UAV thermal imagery data can detect potential thermal anomalies. In this work, we present a machine learning approach for asset extraction of a building’s roof top unit (RTU) using a state-of-the-art object detection algorithm. We also present an approach to identify potential thermal anomalies on the building envelope. Our object detection algorithm achieves 89\% accuracy on the test dataset, while our thermal anomaly algorithms are able to identify potential anomalies, but require further testing for accuracy. The asset information and anomalies are relevant to a variety of urban and energy applications.

A key challenge in determining urban land use is locating industrial regions. The industrial area significantly affects the region's economy and carbon footprints. Most of the time metal roofs identify industries and industrial buildings. We make an effort to use this reasonably typical connection to find industrial buildings. The spectral characteristics of industrial roofs, which have high reflectivity and flat spectrum, serve as the basis for our investigation. With the help of these spectral characteristics of metal roofs, we were able to create an algorithm that is less time-consuming than other methods for target detection, such as matching reference signatures, or every pixel of the image. We have used band ratios for spectral flatness and high reflective characteristics to calculate relative reflectance. The algorithm is tested on the Pavia university dataset, and buildings were detected with a precision of 0.47 which shows the effectiveness of the algorithm. We will use spatial properties like the size to improve the results.

Nighttime light remote sensing data is used to evaluate and measure the effects of human social activities. Gross Domestic Product (GDP) prediction model that is built by long time series nighttime remote sensing data can evaluate and analyze changes of GDP in the relevant area. Chongqing, China was chosen as the study area. NASA's Black Marble nighttime lights product and GDP were modeled to facilitate analysis multiple spatial relationships across the city and its districts. The research results based on the established model show that: there is a strong correlation between NASA's Black Marble nighttime lights product and GDP of Chongqing. The coefficient R2 is 0.9258, 0.9269, 0.9979, respectively, of Linear model, Power exponential model and Logistic model. Among them, Logistic nonlinear regression model has the strongest ability to fit economic data. In this paper, the logistic nonlinear regression model is used to obtain the GDP density map of Chongqing by spatial means, and the development dynamic process and spatial distribution pattern of the GDP change in Chongqing in the past 9 years are analyzed. The research of this paper can provide a new way for the dynamic evaluation of Chongqing's economic data.

This research addressed the important research questions related to the relationship between spatial urban expansion and temperature. Accra was used as test sites by integrating remote sensing, temperature datasets and GIS modelling approaches. The in-depth analysis of spatial urban expansion and temperature was investigated using Land Use Land Cover Changes inventories of 1986, 2003 and 2015. The approaches utilized were adopted to aid investigate the relationship between spatial urban expansion and temperature. It further provided future projections of land use land cover for the year 2025. The results provided empirical evidence that there is a direct relationship between spatial urban expansion and temperature.

Urban green spaces (UGS) play a crucial role in regulating the urban heat island (UHI) effect in cities which has overwhelming urge for expansion and compaction, due to ascending demands of people and activities they accommodate. However, higher rate of population expansion and uncontrolled urban development escalates the concretization drive, especially in cities of developing countries, leading to shrinkage in UGS. This study emphasizes to understand variation in the nature and pattern of Built-up (BU) and UGS for Kolkata city (India) over 30-years (1990 – 2019); and their influence on Land Surface Temperature (LST) which is an important determinant of UHI phenomenon. Time-series Landsat 5, 8 optical bands were used to extract UGS and BU areas using an appropriate Machine Learning (ML) algorithm, and thermal bands were used to retrieve LST. Further a suitable ML algorithm for time-series prediction was employed to determine future UHI scenario of the city in relation to past and existing trend of variability in UGS and BU. The findings indicate towards a positive relationship between decline in UGS density and increase in BU intensity, leading to rise in UHI at a relatively higher rate in the previous two decades which gradually declined in the latter.

The German Research Center for Geosciences also known as GFZ-Potsdam has a long history in the definition and development of new spaceborne sensors such as for gravity and optical Earth Observation missions with GRACE, GRACE-FO, MOMS, and very recently with the launch of EnMAP on 1st April 2022. The Environmental Mapping and Analysis Program (EnMAP) is the first German spaceborne hyperspectral satellite mission. EnMAP aims at monitoring and characterizing the Earth’s environment on a global scale. Core science objectives are toward studying environmental changes, ecosystem responses to human activities, and management of natural resources. The EnMAP mission consortium is composed of the DLR Space Administration in Bonn that is responsible for the overall project management, OHB is responsible of the space segment, DLR Earth Observation Center is responsible of the ground segment, and GFZ Potsdam is responsible for the science related activities and science mission support. In particular, EnMAP is accompanied by an extensive scientific exploitation preparation program that has been run for more than a decade to support industrial and mission development, and scientific exploitation of the data by the user community. In the current EnMAP phase, this program includes mission support during the current commissioning phase and the start of the nominal phase planned toward end of October, supported by the EnMAP Science Advisory Group. In that frame, large activities in the GFZ remote sensing group are dedicated to a) hyperspectral sensor simulation, data quality and validation of EnMAP data products, b) development of methods and open softwares toolboxes such as in the QGIS EnMAP-Box for the pre-processing of radiance to reflectance, and for the retrieval of geo- and bio-physical parameters, c) user community training and workshops, development of new educational resources such as in the EnMAP online learning initiative HYPERedu, opening of a Massive Open Online Course (MOOC) on the basics of imaging spectroscopy, d) mission support and development of validation and background mission plan, and EnMAP announcement of opportunities.

This presentation will give a general introduction to the space debris problem, current state of the environment and currently defined mitigation measures. It will then concentrate on the space debris-related aspects of ESA’s ambitous Space Safety programme and provide details on ESA’s plans to develop sensor technology for debris monitoring in the area of laser, ground- and space-based optical telescopes and radar. In addition to this, the presentation will also address current technology developments towards collision avoidance, space-traffic management and onboard technology to improve European compliance with such requirements in an economically viable way. Finally, the presentation will address the first ever active debris removal mission as an enabler of European industrial capability to conduct in-orbit servicing.

Small-sized on-board radars based on small unmanned aerial vehicles. They have a small aperture of antenna devices, which leads to a wide beam pattern and, accordingly, to a low accuracy in determining the azimuthal coordinates when detecting ground-based physical objects. To improve accuracy under these conditions, multi-position on-board radars are used. In the simplest case, two-position on-board radars. Using data fusion algorithms, they allow to increase the accuracy of determining the azimuthal coordinates by an order of magnitude or more, with a small antenna aperture. In this case, when an object is detected by a two-position radar system, the accuracy estimate is an interval estimate determined by the extreme dimensions of the resolution element of the two-position system. In such a system, the midpoint of the angular azimuthal resolution element in which the physical object is detected is taken as a point estimate. However, other point estimates are possible, which can be strictly computed by specifying the distribution of the object's location in the bin. These point estimates are the mean of the distribution, the bisector, and the median. The proposed astimates in real conditions can be used, for example, for environmental monitoring using appropriate sensors, for assessing the state of the environment, as well as for timely and prompt search and rescue of people in the zone of man-made disasters and disasters in the event of emergencies. At the same time, the key factor is to reduce the search time, which significantly depends on the accuracy of determining the coordinates of objects, which can be radio beacons sending an SOS signal.

Recently, many remote sensing image datasets have been acquired from various platforms, and various semantic segmentation studies using those datasets have been conducted. However, owing to the large data capacity and difficulty of constructing label data, each dataset does not contain several images. Additionally, using all datasets simultaneously is difficult because each dataset has a different spatial resolution, shooting angle, and meaningful objects. In this study, to simultaneously use heterogeneous remote sensing datasets for semantic segmentation, two different unmanned aerial vehicles (UAVs) image datasets, i.e., UAVid semantic segmentation and semantic drone datasets, were used to train a combined universal network (U-Net) model. The flight height of the UAVid semantic segmentation dataset is 50 m above the ground, and it contains 300 images (4096 x 2160 pixels) with 8 classes. Conversely, the flight height of the semantic drone dataset is 5–30 m above the ground, and it contains 598 images (6000 x 4000 pixels) with 20 classes. To minimize the differences between the two datasets, the final classes were adjusted to 8 classes, including building, tree, road, and vegetation. This study confirms the possibility of simultaneously training two different datasets acquired from various places and platforms and evaluates the applicability of semantic segmentation studies using heterogeneous remote sensing datasets.

The paper describes an approach to restoring a three-dimensional model of rigid objects from a single satellite image based on informative classes identified from the results of machine learning, which include railway rails and poles, roofs and walls of buildings, shadows of poles and buildings, and others. The proposed algorithms take into account various conditions for the presence of certain classes in the image, identified by the results of machine learning, as well as the conditions for the absence of metadata on the spatial resolution and spatial orientation of the shooting and the Sun (shooting angle, scanning azimuth, etc.).

We describe the digital technological platform SODIS Building CM for visual control of construction using a BIM model. It can track the work schedule and control the processes of delivery and acceptance of construction and installation works. BIM model elements can be associated with various tasks, documents, equipment, and processes. Elements of the BIM model can automatically change color depending on the state of the object. Staining changes depending on the percentage of completion and the backlog from the base schedule of work. Using a 3D model of an object allows control the amount of work performed on structural elements and all engineering systems. The article also discusses the capabilities of the SODIS Building FM platform, which allows you to get, instead of a pile of operational documentation, which is extremely inconvenient to use, an almost ideal system for safe operation, where the search for the necessary information takes a matter of seconds. This is especially important in the event of an emergency, when there is an acute shortage of time to make effective management decisions.