The establishment of a BES Multi-User Geodatabase (BES-MUG) allows for the storage, management, and distribution of geospatial data associated with the Baltimore Ecosystem Study. At present, BES data is distributed over the internet via the BES website. While having geospatial data available for download is a vast improvement over having the data housed at individual research institutions, it still suffers from some limitations. BES-MUG overcomes these limitations; improving the quality of the geospatial data available to BES researches, thereby leading to more informed decision-making. BES-MUG builds on Environmental Systems Research Institute's (ESRI) ArcGIS and ArcSDE technology. ESRI was selected because its geospatial software offers robust capabilities. ArcGIS is implemented agency-wide within the USDA and is the predominant geospatial software package used by collaborating institutions. Commercially available enterprise database packages (DB2, Oracle, SQL) provide an efficient means to store, manage, and share large datasets. However, standard database capabilities are limited with respect to geographic datasets because they lack the ability to deal with complex spatial relationships. By using ESRI's ArcSDE (Spatial Database Engine) in conjunction with database software, geospatial data can be handled much more effectively through the implementation of the Geodatabase model. Through ArcSDE and the Geodatabase model the database's capabilities are expanded, allowing for multiuser editing, intelligent feature types, and the establishment of rules and relationships. ArcSDE also allows users to connect to the database using ArcGIS software without being burdened by the intricacies of the database itself. For an example of how BES-MUG will help improve the quality and timeless of BES geospatial data consider a census block group layer that is in need of updating. Rather than the researcher downloading the dataset, editing it, and resubmitting to through ORS, access rules will allow the authorized user to edit the dataset over the network. Established rules will ensure that the attribute and topological integrity is maintained, so that key fields are not left blank and that the block group boundaries stay within tract boundaries. Metadata will automatically be updated showing who edited the dataset and when they did in the event any questions arise. Currently, a functioning prototype Multi-User Database has been developed for BES at the University of Vermont Spatial Analysis Lab, using Arc SDE and IBM's DB2 Enterprise Database as a back end architecture. This database, which is currently only accessible to those on the UVM campus network, will shortly be migrated to a Linux server where it will be accessible for database connections over the Internet. Passwords can then be handed out to all interested researchers on the project, who will be able to make a database connection through the Geographic Information Systems software interface on their desktop computer. This database will include a very large number of thematic layers. Those layers are currently divided into biophysical, socio-economic and imagery categories. Biophysical includes data on topography, soils, forest cover, habitat areas, hydrology and toxics. Socio-economics includes political and administrative boundaries, transportation and infrastructure networks, property data, census data, household survey data, parks, protected areas, land use/land cover, zoning, public health and historic land use change. Imagery includes a variety of aerial and satellite imagery. See the readme: http://96.56.36.108/geodatabase_SAL/readme.txt See the file listing: http://96.56.36.108/geodatabase_SAL/diroutput.txt

The National Hydrography Dataset Plus (NHDplus) maps the lakes, ponds, streams, rivers and other surface waters of the United States. Created by the US EPA Office of Water and the US Geological Survey, the NHDPlus provides mean annual and monthly flow estimates for rivers and streams. Additional attributes provide connections between features facilitating complicated analyses. For more information on the NHDPlus dataset see the NHDPlus v2 User Guide.Dataset SummaryPhenomenon Mapped: Surface waters and related features of the United States and associated territories not including Alaska.Geographic Extent: The United States not including Alaska, Puerto Rico, Guam, US Virgin Islands, Marshall Islands, Northern Marianas Islands, Palau, Federated States of Micronesia, and American SamoaProjection: Web Mercator Auxiliary Sphere Visible Scale: Visible at all scales but layer draws best at scales larger than 1:1,000,000Source: EPA and USGSUpdate Frequency: There is new new data since this 2019 version, so no updates planned in the futurePublication Date: March 13, 2019Prior to publication, the NHDPlus network and non-network flowline feature classes were combined into a single flowline layer. Similarly, the NHDPlus Area and Waterbody feature classes were merged under a single schema.Attribute fields were added to the flowline and waterbody layers to simplify symbology and enhance the layer's pop-ups. Fields added include Pop-up Title, Pop-up Subtitle, On or Off Network (flowlines only), Esri Symbology (waterbodies only), and Feature Code Description. All other attributes are from the original NHDPlus dataset. No data values -9999 and -9998 were converted to Null values for many of the flowline fields.What can you do with this layer?Feature layers work throughout the ArcGIS system. Generally your work flow with feature layers will begin in ArcGIS Online or ArcGIS Pro. Below are just a few of the things you can do with a feature service in Online and Pro.ArcGIS OnlineAdd this layer to a map in the map viewer. The layer is limited to scales of approximately 1:1,000,000 or larger but a vector tile layer created from the same data can be used at smaller scales to produce a webmap that displays across the full range of scales. The layer or a map containing it can be used in an application. Change the layer’s transparency and set its visibility rangeOpen the layer’s attribute table and make selections. Selections made in the map or table are reflected in the other. Center on selection allows you to zoom to features selected in the map or table and show selected records allows you to view the selected records in the table.Apply filters. For example you can set a filter to show larger streams and rivers using the mean annual flow attribute or the stream order attribute. Change the layer’s style and symbologyAdd labels and set their propertiesCustomize the pop-upUse as an input to the ArcGIS Online analysis tools. This layer works well as a reference layer with the trace downstream and watershed tools. The buffer tool can be used to draw protective boundaries around streams and the extract data tool can be used to create copies of portions of the data.ArcGIS ProAdd this layer to a 2d or 3d map. Use as an input to geoprocessing. For example, copy features allows you to select then export portions of the data to a new feature class. Change the symbology and the attribute field used to symbolize the dataOpen table and make interactive selections with the mapModify the pop-upsApply Definition Queries to create sub-sets of the layerThis layer is part of the ArcGIS Living Atlas of the World that provides an easy way to explore the landscape layers and many other beautiful and authoritative maps on hundreds of topics.Questions?Please leave a comment below if you have a question about this layer, and we will get back to you as soon as possible.

The research focus in the field of remotely sensed imagery has shifted from collection and warehousing of data ' tasks for which a mature technology already exists, to auto-extraction of information and knowledge discovery from this valuable resource ' tasks for which technology is still under active development. In particular, intelligent algorithms for analysis of very large rasters, either high resolutions images or medium resolution global datasets, that are becoming more and more prevalent, are lacking. We propose to develop the Geospatial Pattern Analysis Toolbox (GeoPAT) a computationally efficient, scalable, and robust suite of algorithms that supports GIS processes such as segmentation, unsupervised/supervised classification of segments, query and retrieval, and change detection in giga-pixel and larger rasters. At the core of the technology that underpins GeoPAT is the novel concept of pattern-based image analysis. Unlike pixel-based or object-based (OBIA) image analysis, GeoPAT partitions an image into overlapping square scenes containing 1,000'100,000 pixels and performs further processing on those scenes using pattern signature and pattern similarity ' concepts first developed in the field of Content-Based Image Retrieval. This fusion of methods from two different areas of research results in orders of magnitude performance boost in application to very large images without sacrificing quality of the output.

GeoPAT v.1.0 already exists as the GRASS GIS add-on that has been developed and tested on medium resolution continental-scale datasets including the National Land Cover Dataset and the National Elevation Dataset. Proposed project will develop GeoPAT v.2.0 ' much improved and extended version of the present software. We estimate an overall entry TRL for GeoPAT v.1.0 to be 3-4 and the planned exit TRL for GeoPAT v.2.0 to be 5-6. Moreover, several new important functionalities will be added. Proposed improvements includes conversion of GeoPAT from being the GRASS add-on to stand-alone software capable of being integrated with other systems, full implementation of web-based interface, writing new modules to extent it applicability to high resolution images/rasters and medium resolution climate data, extension to spatio-temporal domain, enabling hierarchical search and segmentation, development of improved pattern signature and their similarity measures, parallelization of the code, implementation of divide and conquer strategy to speed up selected modules.

The proposed technology will contribute to a wide range of Earth Science investigations and missions through enabling extraction of information from diverse types of very large datasets. Analyzing the entire dataset without the need of sub-dividing it due to software limitations offers important advantage of uniformity and consistency. We propose to demonstrate the utilization of GeoPAT technology on two specific applications. The first application is a web-based, real time, visual search engine for local physiography utilizing query-by-example on the entire, global-extent SRTM 90 m resolution dataset. User selects region where process of interest is known to occur and the search engine identifies other areas around the world with similar physiographic character and thus potential for similar process. The second application is monitoring urban areas in their entirety at the high resolution including mapping of impervious surface and identifying settlements for improved disaggregation of census data.

Detroit Street View (DSV) is an urban remote sensing program run by the Enterprise Geographic Information Systems (EGIS) Team within the Department of Innovation and Technology at the City of Detroit. The mission of Detroit Street View is ‘To continuously observe and document Detroit’s changing physical environment through remote sensing, resulting in freely available foundational data that empowers effective city operations, informed decision making, awareness, and innovation.’ LiDAR (as well as panoramic imagery) is collected using a vehicle-mounted mobile mapping system.

Due to variations in processing, index lines are not currently available for all existing LiDAR datasets, including all data collected before September 2020. Index lines represent the approximate path of the vehicle within the time extent of the given LiDAR file. The actual geographic extent of the LiDAR point cloud varies dependent on line-of-sight.

Compressed (LAZ format) point cloud files may be requested by emailing gis@detroitmi.gov with a description of the desired geographic area, any specific dates/file names, and an explanation of interest and/or intended use. Requests will be filled at the discretion and availability of the Enterprise GIS Team. Deliverable file size limitations may apply and requestors may be asked to provide their own online location or physical media for transfer.

LiDAR was collected using an uncalibrated Trimble MX2 mobile mapping system. The data is not quality controlled, and no accuracy assessment is provided or implied. Results are known to vary significantly. Users should exercise caution and conduct their own comprehensive suitability assessments before requesting and applying this data.

Sample Dataset: https://detroitmi.maps.arcgis.com/home/item.html?id=69853441d944442f9e79199b57f26fe3

A dataset containing the cities of the world

These data show sample locations for various abiotic data collected on Konza Prairie (rain gauges, soil moisture, and stream data). Included in these data are the locations for 12 rain gauges (GIS300) on Konza Prairie. The Konza headquarters weather station formerly consisted of two gauges which were operated year-round. The Konza headquarters weather station currently consists of one Otto-Pluvio2 gauge which is operated year-round. The remaining Konza-operated gauges run from April 1 to November 1. These data are to be used in conjunction with the APT01 (precipitation) dataset. GIS305 defines the locations where measurements of soil moisture (%volume) are taken on Konza Prairie. These data are to be used in conjunction with the ASM01 (soil moisture) dataset. GIS309 defines the locations within watershed N4D of soil sampler nests. In Jan 2020, we separated the original GIS310 file 'Wells in N4D' into GIS310 'Wells in N4D' and GIS309 'Soil Sampler Nests'. Prior to then, soil sampler nests and wells were combined in GIS310. GIS310 defines the locations within watershed N4D where samples are taken for analyzing the belowground water chemistry of the watershed. These data are to be used in conjunction with the AGW01 dataset. GIS311 defines the locations of 14 wells at two sites along Kings Creek. Depth and nutrient content of groundwater is measured at these sites. These data are to be used in conjunction with the AGW02 dataset. GIS315 defines the locations of stream sampling stations within multiple Konza watersheds. These data are to be used in conjunction with the NWC, ASS, ASD, and ASW datasets. GIS320 defines the locations of the rainfall collectors used to collect the samples analyzed as a part of the National Atmospheric Deposition Program. These data are to be used in conjunction with the ANA01 dataset. These data are available to download as zipped shapefiles (.zip), compressed Google Earth KML layers (.kmz).

This dataset is a series of digital map-posters accompanying the AdaptNRM Guide: Helping Biodiversity Adapt: supporting climate adaptation planning using a community-level modelling approach.

These represent supporting materials and information about the community-level biodiversity models applied to climate change. Map posters are organised by four biological groups (vascular plants, mammals, reptiles and amphibians), two climate change scenario (1990-2050 MIROC5 and CanESM2 for RCP8.5), and five measures of change in biodiversity.

The map-posters present the nationally consistent data at locally relevant resolutions in eight parts – representing broad groupings of NRM regions based on the cluster boundaries used for climate adaptation planning (http://www.environment.gov.au/climate-change/adaptation) and also Nationally.

Map-posters are provided in PNG image format at moderate resolution (300dpi) to suit A0 printing. The posters were designed to meet A0 print size and digital viewing resolution of map detail. An additional set in PDF image format has been created for ease of download for initial exploration and printing on A3 paper. Some text elements and map features may be fuzzy at this resolution.

Each map-poster contains four dataset images coloured using standard legends encompassing the potential range of the measure, even if that range is not represented in the dataset itself or across the map extent.

Most map series are provided in two parts: part 1 shows the two climate scenarios for vascular plants and mammals and part 2 shows reptiles and amphibians. Eight cluster maps for each series have a different colour theme and map extent. A national series is also provided. Annotation briefly outlines the topics presented in the Guide so that each poster stands alone for quick reference.

An additional 77 National maps presenting the probability distributions of each of 77 vegetation types – NVIS 4.1 major vegetation subgroups (NVIS subgroups) - are currently in preparation.

Example citations:

Williams KJ, Raisbeck-Brown N, Prober S, Harwood T (2015) Generalised projected distribution of vegetation types – NVIS 4.1 major vegetation subgroups (1990 and 2050), A0 map-poster 8.1 - East Coast NRM regions. CSIRO Land and Water Flagship, Canberra. Available online at www.AdaptNRM.org and https://data.csiro.au/dap/.

Williams KJ, Raisbeck-Brown N, Harwood T, Prober S (2015) Revegetation benefit (cleared natural areas) for vascular plants and mammals (1990-2050), A0 map-poster 9.1 - East Coast NRM regions. CSIRO Land and Water Flagship, Canberra. Available online at www.AdaptNRM.org and https://data.csiro.au/dap/.

This dataset has been delivered incrementally. Please check that you are accessing the latest version of the dataset. Lineage: The map posters show case the scientific data. The data layers have been developed at approximately 250m resolution (9 second) across the Australian continent to incorporate the interaction between climate and topography, and are best viewed using a geographic information system (GIS). Each data layers is 1Gb, and inaccessible to non-GIS users. The map posters provide easy access to the scientific data, enabling the outputs to be viewed at high resolution with geographical context information provided.

Maps were generated using layout and drawing tools in ArcGIS 10.2.2

A check list of map posters and datasets is provided with the collection.

Map Series: 7.(1-77) National probability distribution of vegetation type – NVIS 4.1 major vegetation subgroup pre-1750 #0x

8.1 Generalised projected distribution of vegetation types (NVIS subgroups) (1990 and 2050)

9.1 Revegetation benefit (cleared natural areas) for plants and mammals (1990-2050)

9.2 Revegetation benefit (cleared natural areas) for reptiles and amphibians (1990-2050)

10.1 Need for assisted dispersal for vascular plants and mammals (1990-2050)

10.2 Need for assisted dispersal for reptiles and amphibians (1990-2050)

11.1 Refugial potential for vascular plants and mammals (1990-2050)

11.1 Refugial potential for reptiles and amphibians (1990-2050)

12.1 Climate-driven future revegetation benefit for vascular plants and mammals (1990-2050)

12.2 Climate-driven future revegetation benefit for vascular reptiles and amphibians (1990-2050)

This dataset contains open vector data for railways, forests and power lines, as well an open digital elevation model (DEM) for a small area around a sample forest range in Europe (Germany, Upper Bavaria, Kochel Forest Range, some 70 km south of München, at the edge of Bavarian Alps). The purpose of this dataset is to provide a documented sample dataset in order to demonstrate geospatial preprocessing at FOSS4G2019 based on open data and software. This sample has been produced based on several existing open data sources (detailed below), therefore documenting the sources for obtaining some data needed for computations related to forest accessibility and wood harvesting. For example, they can be used with the open methodology and QGIS plugin Seilaplan for optimising the geometric layout cable roads or with additional open software for computing the forest accessibility for wood harvesting. The vector data (railways, forests and power lines) was extracted from OpenStreetMap (data copyrighted OpenStreetMap contributors and available from https://www.openstreetmap.org). The railways and forests were downloaded and extracted on 18.05.2019 using the open sources QGIS (https://www.qgis.org) with the QuickOSM plugin, while the power lines were downloaded a couple of days later on 23.05.2019.

Additional notes for vector data: Please note that OpenStreeMap data extracts such as forests, roads and railways (except power lines) can also be downloaded in a GIS friendly format (Shapefile) from http://download.geofabrik.de/ or using the QGIS built-in download function for OpenStreetMap data. The most efficient way to retrieve specific OSM tags (such as power=line) is to use the QuickOSM plugin for QGIS (using the Overpass API - https://wiki.openstreetmap.org/wiki/Overpass_API) or directly using overpass turbo (https://overpass-turbo.eu/). Finally, the digitised perimeter of the sample forest range is also made available for reproducibility purposes, although any perimeter or area can be digitised freely using the QGIS editing toolbar.

The DEM was originally adapted and modified also with QGIS (https://www.qgis.org) based on the elevation data available from two different sources, by reprojecting and downsampling datasets to 25m then selecting, for each individual raster cell, the elevation value that was closer to the average. These two different elevation sources are:

• Copernicus Land Monitoring Service - EU-DEM v.1.1 (TILE ID E40N20, downloaded from https://land.copernicus.eu/imagery-in-situ/eu-dem/eu-dem-v1.1; this original DEM was produced by the Copernicus Land Monitoring Service “with funding by the European Union” based on SRTM and ASTER GDEM)
• Digitales Geländemodell 50 m Gitterweite (https://opendata.bayern.de/detailansicht/datensatz/digitales-gelaendemodell-50-m-gitterweite/), produced by the Bayerische Vermessungsverwaltung – www.geodaten.bayern.de –and downloaded from http://www.geodaten.bayern.de/opendata/DGM50/dgm50_epsg4258.tif

This methodology was chosen as a way of performing a basic quality check, by comparing the EU-DEM v.1.1 derived from globally available DEM data (such as SRTM) with more authoritative data for the randomly selected region, since using authoritative data is preferred (if open and available). For other sample regions, where authoritative open data is not available, such comparisons cannot longer be performed.

Additional notes DEM: a very good DEM open data source for Germany is the open data set collected and resampled by Sonny (sonnyy7@gmail.com) and made available on the Austrian Open Data Portal http://data.opendataportal.at/dataset/dtm-germany. In order to simplify end-to-end reproducibility of the paper planned for FOSS4G2019, we use and distribute an adapted (reprojected and resampled to 25 meters) sample of the above mentioned dataset for the selected forest range.

This sample dataset is accompanied by software in Python, as a Jupiter Notebook that generates harmonized output rasters with the same extent from the input data. The extent is given by the polygon vector dataset (Perimeter). These output rasters, such as obstacles, aspect, slope, forest cover, can serve as input data for later computations related to forest accessibility and wood harvesting questions. The obstacles output is obtained by transforming line vector datasets (railway lines, high voltage power lines) to raster. Aspect and slope are both derived from the sample digital elevation model.

World Countries Generalized represents generalized boundaries for the countries of the world as of August 2022. The generalized political boundaries improve draw performance and effectiveness at a global or continental level. This layer is best viewed out beyond a scale of 1:5,000,000.This layer's geography was developed by Esri and sourced from Garmin International, Inc., the U.S. Central Intelligence Agency (The World Factbook), and the National Geographic Society for use as a world basemap. It is updated annually as country names or significant borders change.

This dataset is a geological map of the Windmill Islands, mapped at a nominal scale of 1: 25 000. The map is of lithological units. Structures, etc are ignored. There is a separate, associated, …Show full descriptionThis dataset is a geological map of the Windmill Islands, mapped at a nominal scale of 1: 25 000. The map is of lithological units. Structures, etc are ignored. There is a separate, associated, dataset on geological samples and analyses which has its own metadata record with ID wind_geosamp. A map was produced using this data in February 1997 (see link below).

This world cities layer presents the locations of many cities of the world, both major cities and many provincial capitals.Population estimates are provided for those cities listed in open source data from the United Nations and US Census.

CDFW BIOS GIS Dataset, Contact: Armand Gonzales, Description: These data are the characteristics of the individual snags (standing dead trees) found at 15 sample points with three 0.05-ha circular plot habitat samples taken in 2005 at sample points at Spears and Didion Ranches, Placer County, California. Twelve of the forty-five 0.05-ha circular plots contained snags. To be counted, snags had to be > 4" dbh and > 9.8 ft tall and within the 12.6 m radius plot.

Keymap (World Scale, Bitmap, 6K)

This is a data set of locations at which geological samples were collected at the Windmill Islands. The data is provided as a point shapefile. Attributes of the points include sample number, sample …Show full descriptionThis is a data set of locations at which geological samples were collected at the Windmill Islands. The data is provided as a point shapefile. Attributes of the points include sample number, sample type, lithology and chemical analyses. See the Quality field for further information.

This dataset is one of a number of datasets containing geomorphological data relating to the Windmill Islands, Wilkes Land, Antarctica. The dataset comprises of a digital point coverage of glacial sediments, collected and identified from documented field observations compiled by Dr Ian D Goodwin from his own field notes as well as topographic and surface features identified and interpreted on aerial photographs taken by AUSLIG (now Geoscience Australia) in the 1993-94 field season. Each marine incursion sediment sample is represented as a point which is linked to a separate digital database (ie attribute tables) containing additional information. Sediment samples collected predominantly along the shoreline of these lakes represent a chronology of glacial events/influences on these lakes. Sediment samples taken as a profiled sequence or core can be dated using radiocarbon dating to provide a chronological picture and age occurance of deglaciation and glaciation cycles within the Windmill Islands.

This submission contains a number of maps and shapefiles related to the Utah FORGE site. Examples include geologic maps (several variations) and GIS data for the Utah FORGE site outline. All data are georeferenced to UTM, zone 12N, NAD 83, NAVD 88.

This database was prepared using a combination of materials that include aerial photographs, topographic maps (1:24,000 and 1:250,000), field notes, and a sample catalog. Our goal was to translate sample collection site locations at Yellowstone National Park and surrounding areas into a GIS database. This was achieved by transferring site locations from aerial photographs and topographic maps into layers in ArcMap. Each field site is located based on field notes describing where a sample was collected. Locations were marked on the photograph or topographic map by a pinhole or dot, respectively, with the corresponding station or site numbers. Station and site numbers were then referenced in the notes to determine the appropriate prefix for the station. Each point on the aerial photograph or topographic map was relocated on the screen in ArcMap, on a digital topographic map, or an aerial photograph. Several samples are present in the field notes and in the catalog but do not corresp ...

This dataset is one of a number of datasets containing geomorphological data relating to the Windmill Islands, Wilkes Land, Antarctica. The dataset comprises of a digital polygon/point coverage of …Show full descriptionThis dataset is one of a number of datasets containing geomorphological data relating to the Windmill Islands, Wilkes Land, Antarctica. The dataset comprises of a digital polygon/point coverage of lakes and ponds identified from documented field observations compiled by Dr Ian D Goodwin from his own field notes as well as topographic and surface lake features identified and interpreted on aerial photographs taken by AUSLIG (now Geoscience Australia) in the 1993-94 field season. Each lake/pond is represented as a polygon with a central point label which is linked to a separate digital database (ie attribute tables) containing additional information. Sediment samples collected predominantly along the shoreline of these lakes represent a chronology of glacial events/influences on these lakes. Sediment samples taken as a profiled sequence or core can be dated using radiocarbon dating to provide a chronological picture and age occurance of deglaciation and glaciation cycles within the Windmill Islands.

Types and frequence of public works during the 15th century (10 sample years between 1409 and 1456) in Ghent. GIS layer (linear locations).

This New Zealand Point Cloud Classification Deep Learning Package will classify point clouds into building and background classes. This model is optimized to work with New Zealand aerial LiDAR data.The classification of point cloud datasets to identify Building is useful in applications such as high-quality 3D basemap creation, urban planning, and planning climate change response.Building could have a complex irregular geometrical structure that is hard to capture using traditional means. Deep learning models are highly capable of learning these complex structures and giving superior results.This model is designed to extract Building in both urban and rural area in New Zealand.The Training/Testing/Validation dataset are taken within New Zealand resulting of a high reliability to recognize the pattern of NZ common building architecture.Licensing requirementsArcGIS Desktop - ArcGIS 3D Analyst extension for ArcGIS ProUsing the modelThe model can be used in ArcGIS Pro's Classify Point Cloud Using Trained Model tool. Before using this model, ensure that the supported deep learning frameworks libraries are installed. For more details, check Deep Learning Libraries Installer for ArcGIS.Note: Deep learning is computationally intensive, and a powerful GPU is recommended to process large datasets.The model is trained with classified LiDAR that follows the The model was trained using a training dataset with the full set of points. Therefore, it is important to make the full set of points available to the neural network while predicting - allowing it to better discriminate points of 'class of interest' versus background points. It is recommended to use 'selective/target classification' and 'class preservation' functionalities during prediction to have better control over the classification and scenarios with false positives.The model was trained on airborne lidar datasets and is expected to perform best with similar datasets. Classification of terrestrial point cloud datasets may work but has not been validated. For such cases, this pre-trained model may be fine-tuned to save on cost, time, and compute resources while improving accuracy. Another example where fine-tuning this model can be useful is when the object of interest is tram wires, railway wires, etc. which are geometrically similar to electricity wires. When fine-tuning this model, the target training data characteristics such as class structure, maximum number of points per block and extra attributes should match those of the data originally used for training this model (see Training data section below).OutputThe model will classify the point cloud into the following classes with their meaning as defined by the American Society for Photogrammetry and Remote Sensing (ASPRS) described below: 0 Background 6 BuildingApplicable geographiesThe model is expected to work well in the New Zealand. It's seen to produce favorable results as shown in many regions. However, results can vary for datasets that are statistically dissimilar to training data.Training dataset - Auckland, Christchurch, Kapiti, Wellington Testing dataset - Auckland, WellingtonValidation/Evaluation dataset - Hutt City Dataset City Training Auckland, Christchurch, Kapiti, Wellington Testing Auckland, Wellington Validating HuttModel architectureThis model uses the SemanticQueryNetwork model architecture implemented in ArcGIS Pro.Accuracy metricsThe table below summarizes the accuracy of the predictions on the validation dataset. - Precision Recall F1-score Never Classified 0.984921 0.975853 0.979762 Building 0.951285 0.967563 0.9584Training dataThis model is trained on classified dataset originally provided by Open TopoGraphy with < 1% of manual labelling and correction.Train-Test split percentage {Train: 75~%, Test: 25~%} Chosen this ratio based on the analysis from previous epoch statistics which appears to have a descent improvementThe training data used has the following characteristics: X, Y, and Z linear unitMeter Z range-137.74 m to 410.50 m Number of Returns1 to 5 Intensity16 to 65520 Point spacing0.2 ± 0.1 Scan angle-17 to +17 Maximum points per block8192 Block Size50 Meters Class structure[0, 6]Sample resultsModel to classify a dataset with 23pts/m density Wellington city dataset. The model's performance are directly proportional to the dataset point density and noise exlcuded point clouds.To learn how to use this model, see this story