Map Update Research Articles

AFSNet: attention-guided full-scale feature aggregation network for high-resolution remote sensing image change detection

ABSTRACT Using change detection technique precisely analyzes remote sensing images, it has a broad range of applications in resource surveys, surveillance systems, and map updating. In recent years, deep learning-based methods have become a focus area owing to their excellent feature extraction and representation ability. The fusion of multi-scale features is the key to improving change detection performance in fully convolutional network-based structural methods, primarily based on an architecture with skip connections or nested and dense skip connections. However, these methods only fuse features on the same scale and lack sufficient information from multiple scales to generate appropriate results. Traditional feature fusion with redundant and unsupervised information leads to poor model fitting. To solve these problems, we proposed a novel attention-guided full-scale feature aggregation network (AFSNet). The proposed method used a Siamese structure as the backbone network to extract features, which were then aggregated using full-scale skip connections, and an attention mechanism to avoid feature redundancy. Finally, to obtain a highly accurate final change map, a multiple side-outputs fusion strategy was used to fuse the change maps at different scales. To check the reliability of AFSNet, we tested it on two public datasets, the LEVIR-CD and SVCD datasets. The F1-Score/IoU scores improved by 0.57%/0.95% and 1.14%/2.07% in the two datasets, respectively, compared to those obtained using the methods that achieved suboptimal values. The results showed that AFSNet outperforms other mainstream methods while maintaining a good balance between computational costs and model parameters.

Correction, update, and enhancement of vectorial forestry road maps using ALS data, a pathfinder, and seven metrics

Accurate information about forestry roads is a key aspect of forest management in terms of economy (e.g. accessibility, cost, optimal path) and ecology (e.g. wildfire and wildlife protection). In Canada, and in fact, globally, most provincial, state or territory governments maintain vectorial information on the forestry roads under their jurisdiction. However, official maps are not always accurate, may lack road attributes of interest and are not always up-to-date. Airborne Laser Scanning (ALS) has become an established technology to accurately characterize and map broad territories by providing high density 3D point-clouds with, at least, 3 or 4 measurements per square meter. This paper addresses the problem of the automatic updating, fixing, and enhancement of vectorial forestry road maps over large landscapes (¿10000 km 2 ). For this purpose, we developed a production ready, documented and open-source software. From metrics derived from the point-cloud the method produces a raster of road probability. It then uses an existing, inaccurate, map of the road network to define approximate start and end points for each road. Then, a pathfinder retrieves the accurate road shape by computing the least cost path between the two points on the probability raster. Using the accurate road position given by the algorithm, road width and road state are then estimated based the on characteristics of the point-cloud. We demonstrate that our algorithm retrieves the centrelines of roads in a natively vectorial form with an error below 3 m in 95% of the roads using a fully automatic method. The accuracy of the road location allows us to derive other accurate measurements, including the state of the roads. • We accurately relocated roads from governmental database using ALS point-clouds. • We measured the width and estimated the state of each road segment. • The method is natively vectorial and uses pathfinding instead of pixel classification. • The method is open-source, fully packaged, documented, and production-ready.

NanoMap: A GPU-Accelerated OpenVDB-Based Mapping and Simulation Package for Robotic Agents

Encoding sensor data into a map is a problem that must be undertaken by any robotic agent operating in unknown or uncertain environments, and real-time updates are crucial to safe planning and control. Most modern robotic sensors produce some form of depth data or point cloud information that is only useful to the agent after being processed into the appropriate data structure, oftentimes an occupancy map. However, as the quality of sensor technology improves, so does the magnitude of the input data, which can creates a problem when trying to construct occupancy maps in real-time. Populating such an occupancy map using these dense point clouds can quickly become an expensive process, and many robotic agents have limited onboard computational bandwidth and memory. This results in delayed map updates and reduced operational performance in dynamic environments where real-time information is crucial for safe operation. However, while many modern robotic agents are still relatively limited by the power of onboard central processing units (CPUs), many platforms are gaining access to onboard graphics processing units (GPUs), and these resources remain underutilised with respect to the problem of occupancy mapping. We propose a novel probabilistic mapping solution that leverages a combination of OpenVDB, NanoVDB, and Nvidia’s Compute Unified Device Architecture (CUDA) to encode dense point clouds into OpenVDB data structures, leveraging the parallel compute strength of GPUs to provide significant speed advantages and further free up resources for tasks that cannot as easily be performed in parallel. An evaluation of our solution is provided, with performance benchmarks provided for both a laptop and a low power single board computer with onboard GPU. Similar performance improvements should be accessible on any system with access to a CUDA-compatible GPU. Additionally, our library provides the means to simulate one or more sensors on an agent operating within a randomly generated 3D-grid environment and create a live map for the purposes of evaluating planning and control techniques and for training agents via deep reinforcement learning. We also provide interface packages for the Robotic Operating System (ROS1) and the Robotic Operating System 2 (ROS2), and a ROS2 visualisation (RVIZ2) plugin for the underlying OpenVDB data structure.

Road extraction through digital processing and visual interpretation of satellite images

Road extraction plays an important role in urban planning and city extension issues, as well as in road monitoring, traffic management and map updating. Technological advances may offer a wealth of data and techniques that could be implemented in road delineation and extraction, as well as in change detection projects. Among the various methods that have been developed for this purpose, remote sensing techniques and especially digital processing of satellite data could contribute significantly in this direction. This paper presents the study of a road network which concerns the city of Kastoria located in northwestern Greece and its surrounding area. The special character of the road network is closely related to the city’s cultural value which is depicted in its structure. Satellite images of various spatial resolutions were employed for digital processing. Landsat 8 imagery was used in order to detect and delineate the linear features through spatial enhancement, while Landsat 5 images were used to detect changes over time through visual interpretation. Semi-automatic techniques were applied to SPOTmaps products to extract the road network, while an object-oriented approach was applied to QuickBird imagery in order to combine the spatial components with spectral properties. Through semi-automatic digitization and object-oriented workflows, the export of the studied part of the road network in vector format was achieved, thus facilitating the process and reducing the required time. The resulted data are efficient in road network delineation and could be combined with other data for road maintenance and extension, change detection issues, as well as for cultural and touristic purposes.

C C I B A*: An Improved BA* Based Collaborative Coverage Path Planning Method for Multiple Unmanned Surface Mapping Vehicles

The main emphasis of this work is placed on the problem of collaborative coverage path planning for unmanned surface mapping vehicles (USMVs). As a result, the collaborative coverage improved <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$BA^{*}$ </tex-math></inline-formula> algorithm ( <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$CCIBA^{*}$ </tex-math></inline-formula> ) is proposed. In the algorithm, coverage path planning for a single vehicle is achieved by task decomposition and level map updating. Then a multiple USMV collaborative behavior strategy is designed, which is composed of area division, recall and transfer, area exchange and recognizing obstacles. Moverover, multiple USMV collaborative coverage path planning can be achieved. Consequently, a high-efficiency and high-quality coverage path for USMVs can be implemented. Water area simulation results indicate that our <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$CCIBA^{*}$ </tex-math></inline-formula> brings about a substantial increase in the performances of path length, number of turning, number of units and coverage rate.

Lithostructural Mapping Using Landsat OLI images and Field Investigations in the Oumjrane–Boukerzia Mining District, Eastern Anti-Atlas, Morocco

The application of remote sensing is considered to be highly efficient in the field of geology, particularly for mapping and discrimination between lithological units, as well as the identification of different surface minerals which enable the acquisition of all the optical and radar data needed to obtain more reliable and particular information, especially in inaccessible areas. Landsat 8 (Operational Land Imager) image bands, which include color compositions 7, 4, and 1, band ratios 7/5, 4/2, 3/1, and 6/3, 6/1, 5/2, Principal Component Analyses PC1, PC4, and PC3, and directional filtering at 0° (NS), 45° (NE–SW), and 90° (EW), were used to analyze the spectral characteristics of the lithological units of the study area and delineate the alteration zones which may contain significant concentrations of base metals. Combined with geological field observations, these data were integrated and analyzed in a geographic information system to establish a multi-criteria characterization of mineralized zones. Hence, the use of remote sensing contributed to the improvement and updating of available geological maps of the Oumjrane–Boukerzia area. Lineament extraction allowed for performed analyses of the structural elements that play a crucial role in the mineral distribution in the study area. Several of the targeted areas were examined in the field by selective sampling. Consequently, the results helped identify new mineralized zones with important mining potential which could be evaluated in the future through detailed geophysical and geochemical work.

Semi-automatic road extraction from high resolution satellite images by template matching using Kullback–Leibler divergence as a similarity measure

ABSTRACT Semi-automatic extraction of roads is greatly needed to accelerate the acquisition and updating of road maps. However, road surfaces are frequently disturbed on very high spatial resolution (VHSR) remotely sensed satellite imagery, which bothers the road trackers using least-squares-based template matching. This paper presents a novel semi-automatic framework for road tracking from VHSR satellite imagery. First, a human operator inputs three seed points. Second, the computer automatically tracks the road by the template matching using Kullback–Leibler divergence as a similarity measure. At the same time, a human operator is retained in the tracking process to supervise the extracted results, to response to the program’s prompts. Once the failure or error happens, the human operator will correct the results and restart the automatic tracking. The above procedure is repeated until a whole road is tracked. Four satellite images with different complexities are used to perform experiments. The results show that our proposed road trackers is capable of automatically, accurately and fast extracting the long and high-level roads from the VHSR satellite images.

CrowdLoc: Robust image indoor localization with edge-assisted crowdsensing

Image indoor localization, which is advantageous in infrastructure-less deployment and high positioning accuracy among various indoor localization schemes, is usually deployed in a lightweight end (shooting picture) and heavyweight server (searching image for location in the database) manner. Such a scheme is suffered from two major drawbacks: (1) The uploading and searching of the shooting pictures cause a severe burden to the server. (2) The image map will become defected as time elapse, such as missing features in initialization maps and continuous changes in the scene. To keep the image map up-to-date, a labor-intensive site survey is required with dedicated devices. The frequent updating will further exhaust the limited resources in the server. These two drawbacks have greatly hindered the image-based indoor localization from large-scale deployment. To this end, we study how to use crowdsourcing data to mitigate map defects and optimize location computation and map storage through edge architecture. In this paper, we propose CrowdLoc, an edge-assisted image localization architecture with lightweight multi-view localization on clients and crowdsourcing map-cache on the edge. Our method is mainly innovative in three parts. Firstly, CrowdLoc provides a lightweight multi-view localization on the client-side and the low-confidence localization can be offloaded to the edge server to use the crowdsourcing map for more accurate calculation. Secondly, CrowdLoc uses a map defect perception algorithm to recognize the update of the scene during user localization and update the crowdsourcing map-cache automatically on the edge side. Thirdly, in order to improve cache utilization, we propose a cache elimination strategy that runs periodically on the edge side. It can classify cache data according to the frequency of cache hits on a regular basis, remove invalidation data, and persist long-term updated data to the client map to reduce clients’ computing offload. We evaluate the performance of our method based on the real data in the lobby of an experimental building with a map missing rate of 20% and a map update rate of 40%. The results show that the mean localization error of CrowdLoc is 1.54 meters, which is 59.4% lower than that of multi-view localization without crowdsourcing maps, and only 0.56 meters higher than localization in non-defective maps.

Target Search in Dynamic Environments With Multiple Solar-Powered UAVs

This article focuses on target search in dynamic environments using multiple solar-powered unmanned aerial vehicles (SUAVs) with the probability map method. First, the energy model of the SUAV and an external solar environment model are established. Second, the Gaussian mixture model (GMM) is used to extract the prior probability distribution of the target to obtain multiple search areas composed of two-dimensional Gaussian distributions. Then, dynamic task allocation of the SUAVs based on the environment is performed utilizing the GMM information. Next, the SUAVs search for the target according to the assignment and plan a path with the energy model and probability established before, and the probability map is updated in real time according to the detection result of the SUAVs and the velocity vector field of the environment. In this study, two probability map update methods, namely the equivalent area iteration (EAI) and probability estimated diffusion (PED) methods, are proposed for the situations of precise and imprecise environmental information, respectively. Finally, comparative simulations are carried out for these two different situations to prove the improvement obtained using SUAVs.

ON THE APPLICATION OF EFFECTIVE METHODS OF ORGANIZATION OF GEODETIC AND CARTOGRAPHIC WORKS IN THE CZECH REPUBLIC

The purpose of the study was determined by new requirements and significant tasks of the promising development of the cartographic and geodetic industry of the country, which were designated as the development of scientific foundations for the use of the latest technologies of system mapping and cartographic modeling, the search for new types, types of maps that would thoroughly reproduce the relationship and dynamics of natural and socio-economic phenomena, and the like. The use of GNS and GIS technologies, remote sensing of the Earth, will certainly help to accelerate the solution of problems related to the creation and updating of topographic maps and plans. The use of different coordinate systems, an excellent mapping system and nomenclature makes it difficult to solve them. At the same time, modern technologies, the study and application of the experience of specialists from other countries will make it possible to carry out all these processes efficiently and rationally.&#x0D; The article describes the experience of using effective methods of organization of geodetic and cartographic work abroad on the example of the analysis of work in the Czech Republic.&#x0D; The set goal and objectives of the research led to the use of general scientific approaches, logical laws of drawing conclusions, and special methods of cognition. In carrying out the tasks, the research was based on the world experience of using the descriptive method, the comparative historical method (comparativism), the historical and typological method based on complex and systematic approaches. The information base of the research was collections of scientific papers, periodical professional publications, Internet resources, regulatory documents.

On the Accuracy of Cadastral Marks: Statistical Analyses to Assess the Congruence among GNSS-Based Positioning and Official Maps

Cadastral marks constitute a dense source of information for topographical surveys required to update cadastral maps. Historically, in Italy, cadastral marks have been the cartographic network for the implementation of mapping updates. Different sources of cadastral marks can be used by cadastral surveyors. In recent years, the cadastre is moving toward a digital world, and with the advancement of surveying technology, GNSS CORS technology has emerged in the positioning of cadastral marks. An analysis of congruence among cadastral marks using GNSS CORS and official maps is missing. Thus, this work aims to analyze the positional accuracy of some cadastral marks, located in Palermo, Italy, with regard to the official maps produced by the cadastral bureau, the local cartography, and Google Earth maps. A survey of 60 cadastral marks was carried out by conventional GNSS NRTK procedures, with the lateral offset method due to their materialization (mostly building edges), which is not always directly detectable. The cadastral marks’ positioning was obtained from different maps: cadastral maps and related monographic files, numerical technical maps, and Google Earth maps, to check their coordinate congruence. A statistical approach was performed to check whether the distribution frequencies of the coordinate’s differences belonged to the bivariate normal distribution for the planimetric coordinates and the univariate normal distribution for the altimetric component. The results show that the hypothesis of a normal distribution is confirmed in most of the pairs, and specifically, most of the analyses indicate that the highest congruencies seem to characterize the coordinates determined by using the GNSS and with those that can be deduced by the numerical technical maps. The results obtained experimentally show centimetric accuracies obtained by the GNSS NRTK survey, in both the planimetric and altimetric components, while the accuracies obtained from the georeferencing of the cadastral maps show differences in the order of 0.4–0.8 m. Meanwhile, the differences resulting from comparing the technical cartography produced by the local authority and Google Earth maps show greater criticalities, with a metric order of magnitude.

Dynamic modeling for noise mapping in urban areas

Environmental noise has been a major environmental nuisance in metropolitan cities. To achieve the goal of sustainable community, noise reduction is an important approach. Without systematic noise mapping, the spatio-temporal distribution of noise variations is hard to capture. This study proposes a new methodology framework to combine statistical models and acoustic propagation for dynamic updates of 2D and 3D traffic noise maps by using a limited number of noise sensors in Taipei City based on multisource data including noise monitoring, vehicle detectors, meteorological data, road characteristics, and socio-demographic data. The hourly mean difference between the predicted and measured noise level is within the range of −6.25 dBA to −4.46 dBA in the 2D noise model. For the 3D noise model, the hourly mean prediction error is within the range of 0.02 dBA to 1.93 dBA. Based on the WHO benchmark for excessive road traffic noise, we found at least 30% of inhabitants in Taipei City are exposed to levels exceeding 53 dBA Lden, and >25% are exposed to noise levels exceeding 45 dBA Lnight. The noise maps not only can help identify vulnerable communities to adopt proper approaches for noise reduction but also can remind the residents to take action to improve their quality of life.

A Tight Filtering and Smoothing Fusion Method With Feature Tracking for LiDAR Odometry

LiDAR odometry (LO) has gained popularity in recent years because of its growing applications. Existing solutions include filtering and smoothing, where the former optimizes the latest pose with high efficiency and the latter optimizes multiple poses with strong consistency. Still, it is promising to fuse filtering and smoothing in LO. In this paper, we propose a novel tight filtering and smoothing fusion method. Firstly, two nested sliding windows are maintained and they share estimated poses and features to balance the accuracy and efficiency. The large outer window is used to build a large local map for high-quality scan-to-map matching in the filtering module, and multiple poses in the small inner window are optimized in the smoothing module to improve local consistency and efficiency. Furthermore, sparse directed geometry points (DGPs) are tracked across window keyframes to produce feature-wise multiple residuals for poses optimization in both filtering and smoothing modules. Besides, the strategies of ratio-based refined pose optimization and efficient DGP feature triangulation are developed for pose refinement and local map update, respectively. The experiments on KITTI and Newer College (NC) datasets demonstrate the effectiveness of the proposed method.

IMPROVED METHODOLOGIES FOR THE REVISION OF A TRADITIONAL TOPOGRAPHIC MAP SERIES

Abstract. Conventional topographic maps are still important for many applications, such as land planning, but require frequent updates, which involve high costs and are not easy to accomplish. There are however many advances in Photogrammetry that are not always exploited to their full potential in a real map production environment. The traditional process involves aerial triangulation with ground control points, stereo restitution in digital photogrammetric workstations and field completion. The process is very time-consuming and expensive and map update is many times not done with the frequency required by users.This paper describes the implementation of an improved methodology for very frequent revision of the Portuguese map series of scale 1:25,000 making use of conventional digital aerial photography, which now is acquired on a regular basis for all the country. Vector features are digitised in monoscopic mode, in a GIS environment, over true orthophotos. These can be produced in a largely automated manner, using direct georeferencing data, and with minimal ground control only for vertical adjustments. Revision time and cost can be significantly reduced, keeping acceptable accuracy standards. That will satisfy the users that require the traditional high standards of the topographic map series for land panning.Additionally, for the task of field data collection of information that cannot be obtained from aerial photographs, a method is being applied to make use of video data collected in the field along roads with action cameras, with GPS. Operators can confirm or identify attributes of objects along, or near roads, such as power lines.

Updated spatial data of detailed topographic monitoring is the basis for the rapid updating of state digital topographic maps: methods and technologies

A modern solution to the problem of modernization of the state system of cartographic provision of the Russian Federation with up-to-date spatial data is the creation of a system of state topographic monitoring (GTM system). The purpose of the study is to develop theoretical aspects, methods and technological techniques: conducting detailed topographic monitoring (DTM); forming a digital topographic map on duty (original changes), a database of changes in terrain objects; integrating DTM data into the process of updating state topographic maps. The results of the study are scientifically substantiated, can be regulated and practically tested. The integration of the proposed scientific, methodological and technical solutions in the process of updating the state digital topographic maps (CTM) will ensure: maintaining the cartographic fund on the territory of the country and on the border areas at the level of modernity; the efficiency of updating the state CTK.

A new coverage path planning algorithm for unmanned surface mapping vehicle based on A-star based searching

This paper investigates the coverage path planning issues and mapping platform construction for unmanned surface mapping vehicle (USMV). An improved BA∗ (IBA∗) algorithm is designed from the unit decomposition method and map update method to solve the problem of insufficient continuity and the high-precision environmental modeling. By means of task decomposition and map dynamic updating, the IBA∗ algorithm overcomes the shortcoming of local optimization of BA∗ algorithm effectively, and has a significant decrease more than 10% in the path length, 15% in the number of turns, 85% in the unit number and 2% in the coverage. Moreover, the IBA∗ algorithm has a greater advantage than the Boustrophedon algorithm and the PPCPP algorithm. Results of simulations and the mapping experiments verify the excellent performance of our IBA∗.

Cloud Update of Geodetic Normal Distribution Map Based on Crowd-Sourcing Detection against Road Environment Changes

LiDAR-based localization has been widely used for the pose estimation of autonomous vehicles. Since the localization requires a sustainable map reflecting environment changes, a map update framework based on crowd-sourcing measurements has been researched. Unfortunately, a point cloud map occupies too large data size to transmit data in the uploading and downloading of the map update framework. To realize the LiDAR map update framework by reducing the data size, we proposed a novel map update framework using a Geodetic Normal Distribution (GND) map that compresses the point cloud to the normal distributions. The proposed GND map update framework comprises two parts: map change detection based on crowd-sourcing vehicles and map updating based on a map cloud server. GND map changes are detected based on an evidence theory considering geometric relationships between the GND map and crowd-sourcing measurements and uploaded to the map cloud server. Uploaded map changes reproduce representative map changes based on a similarity-based clustering, which are updated into the GND map. The proposed framework was evaluated in simulations and real environments on construction sites. As a result, although partial map changes occurred, the GND map was kept up-to-date through the proposed framework and the localization for autonomous driving was performed successfully.

UFOExplorer: Fast and Scalable Sampling-Based Exploration With a Graph-Based Planning Structure

We propose UFOExplorer, a fast and efficient exploration method that scales well with the environment size. An exploration paradigm driven by map updates is proposed to enable the robot to react quicker and to always move towards the optimal exploration goal. For each map update, a dense graph-based planning structure is updated and extended. The planning structure is then used to generate a path using a simple exploration heuristic, which guides the robot towards the closest exploration goal. The proposed method scales well with the environment size, as the planning cost is amortized when updating and extending the planning structure. The simple exploration heuristic performs on par with the most recent state-of-the-art methods in smaller environments and outperforms them in larger environments, both in terms of exploration speed and computational efficiency. The implementation of the method is made available for future research.

Accuracy Issues for Spatial Update of Digital Cadastral Maps

All geospatial data are updated periodically. Cadastral parcel mapping, however, has special update requirements that set it apart from other geospatial data. Mapped boundaries change continuously to fit with new survey plans. Additionally, new parcels have to be fitted and aligned with adjoining parcels to merge them into existing cadastral mapping. This is preferably performed by a spatial adjustment approach to systematically improve its accuracy over time. This paper adapts methods for analysis and adjustment of survey networks to improve the accuracy of cadastral mapping with better coordinate positioning and survey plan dimensions. Case studies for both hypothetical and real cadastral mapping are used to illustrate the issues and spatially resolve errors. Adjustment results achieve an accuracy consistent with other GIS layers and boundary features visible in high-resolution orthoimagery. Graphical charts based on stress–strain relationships provide a simplified means to interpret post-adjustment results to identify and fix potential errors.

Automatic extraction of land cover statistics from satellite imagery by deep learning1

In this paper, we address the challenge of producing fully automated land cover estimates from satellite imagery through Deep Learning algorithms. We developed our system according to a tile-based, classify-and-count design. To implement the classification engine of the system, we adopted a cutting-edge Convolutional Neural Network model named Inception-V3, which we customized and trained for scene classification on the EuroSAT dataset. We tested and validated our system on two Sentinel-2 images representing quite different Italian territories (with an area of 751 km2 and 443 km2, respectively). Because no genuine ground-truth is available for the land cover of these sub-regional territories, we built a pseudo ground-truth by integrating land cover information from flagship European projects CORINE and LUCAS. A critical and careful analysis shows that our automatic land cover estimates are in good agreement with the pseudo ground-truth and offers extensive evidence of the remarkable segmentation ability of our system. The limits of our approach are also critically discussed in the paper and possible countermeasures are illustrated. When compared with traditional projects like CORINE and LUCAS, our automatic land cover estimation system exhibits three fundamental advantages: it can dramatically reduce production costs; it can allow delivering very timely and frequent land cover statistics; it can enable land cover estimation for very small territorial areas, well beyond the NUTS-2 level. As an additional outcome of land cover estimation, our system also automatically generates moderate resolution land cover maps that might be used in cartography projects as an agile first-level tool for map update or change detection purposes.