“20 million address points to postal code polygons – The evolution of the Canadian Postal Code Crowdsourced Dataset.”
Ervin Ruci;
Talk
I've been compiling a crowdsourced postal address file for Canada since 2005 as a derivation of the geocoder.ca API raw data. I'll show how I'm using human input and machine learning to compute Canadian Postal Code Polygon boundaries based on an evolving dataset of over 20 million input points.
“A Practical Guide to Applying Computer Vision Techniques to Geospatial Imagery”
Chris Brown, Joe Morrison;
Talk
Deep learning and other computer vision techniques are revolutionizing the usage of imagery in application development. How well do these techniques transfer to geospatial imagery? Geospatial imagery is unique in its metadata and size compared to many of the traditional applications of these technologies. This talk describes the leading computer vision techniques and how they apply to geospatial imagery. Attendees will learn when these techniques can be used with geospatial imagery and the open source tools and standards can be used. Additionally, this talk covers which geospatial projects lend themselves to deep learning techniques and how to measure success.
“Bayanihan Maps: A project for Solid Waste Management in the Philippines”
Ramon Ayco Jr., Sigfred B Balatan Jr;
Talk
Smartphones and computer screens have become our primary “window” to the world, and no matter how much mess we have in the “real world” we don’t often talk about it online. This is addicting because we seem to have control over what our screens show us.
Our project aims to capture the country’s solid waste management problems in a fun way and bring this to people’s attention. The hope is to spark conversations and aid in creating policies that might bring people to action and even help prevent future hazards through open-source software and open data.
“Building a modern geospatial IoT platform at scale using open-source software, open standards, and open data”
Abraham Poorazizi;
Talk
The IoT devices, from location-tracking solutions to wearable sensors, generate a massive amount of geospatial data in real-time at an astounding rate. This phenomenon imposes a set of unique challenges on geospatial IoT platforms in terms of data ingestion, processing, and visualization, which are necessary steps to derive insights from the data and improve the decision-making processes. In this presentation, I discuss how we, at SensorUp, are building a modern and scalable geospatial IoT platform using an open ecosystem. I also demonstrate how such a platform can solve real-world problems across different industries.
“Building an business using FOSS4G”
blair freebairn;
Talk
I will share the journey of Geolytix. A FOSS4G business that doesn't talk much about open source but relies on it every day. Geolytix was founded eight years ago and has grown into a hybrid product/services business with 35 geospatial experts. We are a mix of data scientists, developers, data engineers and location planners. We tell people where to go, primarily large retailers.
“Building an off-grid solar GIS in West Africa”
Alex Orenstein;
Talk
Rural electrification in West Africa remains low. In many countries, less than 20% of the rural population is on the electrical grid. As a result, off-grid home solar power is rapidly growing in West Africa and a need for geodata and GIS products adapted to the unique context of the sector. Oolu, a Senegalese off-grid solar company is building a GIS to both manage operations and estimate off-grid populations. This talk will outline the system’s open-source nuts and bolts, the methodology behind estimating off-grid households and what we learned in the process.
“Building Digital Earth Africa”
Alex Leith;
Talk
In 2019, Geoscience Australia announced the creation of the ambitious Digital Earth Africa initiative, modelled on the rising success of Digital Earth Australia. The goal of the Digital Earth platforms is to make petabytes of Earth observation data freely available and accessible to inform policy, stimulate economic growth, and build a deeper understanding of our dynamic planet. This talk will discuss what the goals of Digital Earth Africa are and how the technology empowers African people.
“Case study of data storage for preservation of our archiving system at the National Geographic and Hydrographic Institute of Madagascar”
BAOVOLA;
Talk
We are storing and backing up with the aim of saving our national heritage data (vector and raster databases, cartographic and geodetic works, old photography); whatever their nature and their physical supports. In fact, our budget is limited in terms of software license, so we are using FOSS4G to make our organization better with QGIS and PostgreSQL/PostGIS to migrate from Shapefiles to rows. Our methodology might be elementary but we believe, not only we would like to share our experiences; but also developing countries might have same problematic as us regarding to how to preserve their old heritage data.
“CensusMapper and {cancensus}, putting the "open" into StatCan Open Data”
Jens von Bergmann;
Talk
CensusMapper is an open platform to allow anyone to easily create interactive maps based on Canadian census data, share those maps, and browse maps created by others. It also implements an API server for targeted access to census data for research, facilitated further by the {cancensus} R package.
This talk provides an overview of these open data tools and what's we are hoping to accomplish in the future.
“Community generated data for informed decisions in rapidly urbanizing cities, a case study of Dar es salaam”
Hawa Adinani;
Talk
In Tanzania, developing the government’s response capacity to use the data is a major challenge, as the agencies responsible do not necessarily have the know-how to utilize all datasets. Changing ways of government decision making that have existed over the years is a further challenge.
Ramani Huria, a community mapping project for flood resilience in Dar es salaam has been working closely with community members from flood-prone areas to provide vital information and accurate datasets. Including getting proper recommendations from them on what should be done to reduce the severity of the disaster by adopting sustainable solutions.
“Deep Learning based Cloud Mask for Satellite Images at Scale”
Pramithus Khadka;
Talk
Clouds in satellite imagery result in biased vegetation indices, erroneous land cover classifications, and inaccurate biomass inferences. The Climate Corporation’s deep learning based cloud mask outperforms traditional machine learning based cloud mask for multiple imagery sources and can be applied efficiently at scale.
“Elevation in the Open: Migrating USGS 3D Elevation Program Products to Open File Formats”
William Lane;
Talk
In 2018, the US Geological Survey 3D Elevation Program (3DEP) began an effort to migrate all elevation products (point cloud/raster/vector) from legacy or proprietary formats to cloud-ready, open standard formats. We completed migration of all newly published products to open formats (LAZ, COG, Geopackage) in 2019, and will finish migrating existing products in 2020. This presentation will review the history of formats used by 3DEP (and formerly the National Elevation Dataset), the challenges of format migration for a national program with significant amounts of legacy data, and benefits achieved as a result of utilizing open data standards.
“Fight Club: Sentinel Versus Planet”
Desire Kaczynski Rukud'de, Evet Naturinda;
Talk
2014 marked the time when the EU took the title from NASA when they launched the Copernicus Program and their network of upgraded satellites. NASA’s LandSat held that rank for decades with the LandSat Program which was much loved by scientist groupies everywhere. Recently, however, some VC backed new-kids-on-the-block have begun to shake things up in the space space with their unorthodox methods; shoe box size satellites, low cost and quick turnaround images. Can this scrappy newbie really compete at this level? Do they have what it takes?
“geocube - tool to convert geopandas vector data into rasterized xarray data”
Alan Snow;
Talk
In digital agriculture work, data scientists combine a variety of both vector and raster datasets to produce new insights in the form of soil maps, machine learning models, and advanced visualizations. To effectively marry up vector and raster data, you need to rasterize the vector data in context-dependent ways. We have simplified this process with an open-source Python library called geocube. Now you can move data from GeoDataframes (geopandas) and other vector file formats into xarray Datasets with a variety of advanced rasterization options. Once in xarray, you can export to standard raster or multidimensional data formats.
“Geo in Insurance: from back-office to the front-line”
Mark Varley;
Talk
Location is important in all types of insurance and for property insurance it is the key risk factor. It is therefore no surprise that insurers were some of the earliest adopters of GIS. In the digital age however traditional GIS enterprise architectures no-longer scale as geo is going mainstream. In this talk we will explore some key insurance use cases from risk assessment to exposure management and how modern geo services are transforming the insurance business.
“Geospatial Indexing in Apache Lucene and Elasticsearch”
Nicholas Knize, PhD;
Talk
Come have a look under the covers at new data structures that enable geospatial and multi-dimensional indexing and search at massive scale in Apache Lucene and Elasticsearch. This talk will cover not only the indexing structures considered and ultimately implemented in the Apache Lucene Open Source Project but the exceptional performance improvements and centimeter spatial accuracy obtained in the 7.6 and 8.0 release. As a bonus, this talk will cover new and upcoming Spatial Analysis Aggregations and Processing available in the Elastic Stack.
“Hack4Equity: Vocalizing Gender Based Stats in East Africa”
Yariwo Kitiyo, Caroline Akoth Otiwa;
Talk
Women in GIS Kenya is an initiative that mentors and builds innovative solutions to africa's most pressing gender issues. As a cofounder and CEO of the initiative, I have been spearheading the use of data to address issues that are faced by minority groups. Our major milestone was the violence week where we created info graphics around GBV across Africa to create awareness and get the attention of policymakers.
“Make Vector Tiles More Popular than Shapefiles with Your Colleagues”
Max Malynowsky;
Talk
"Can you send me that as a shapefile?" First introduced in the '90s, it refuses to go away as the de-facto interchange format among casual GIS users, no matter how much we want to deprecate it. On the web, vector tiles have assumed that ubiquitous standard, with open and closed systems agreeing to implement it in the same way (mostly). What's more, integrations with desktop software has matured, and their convenience might be enough to convince people to prefer it. All this is possible without having to run servers or lock down your data to a specific platform.
“Mapping for Safety and Crime in the Informal settlement”
Nicera Wanjiru Kimani;
Talk
Personal Security in the informal settlements in cities like Nairobi is a major concern for residents our informal settlements are full of crime. with help of FOSS4G networks i can now help my community solve the problems that we have been facing.
This paper explores how the use of aerial survey data and maps can be applied by local government, community managers and mappers to better identify and reduce public risk in a proactive way and also it evaluates safety concerns and crimes that take place within our communities and initiatives that can be adopted to reduce societal crime.
“On Demand Machine Learning Pipelines with STAC/COG”
Jeff Albrecht;
Talk
Arturo provides on-demand physical property characteristics and predictions for residential and commercial properties. This talk will discuss how Arturo leverages the scaleability of Cloud Optimized GeoTiffs (COG) and the interoperability of STAC to build an event-driven data pipeline which feeds the most current satellite, aerial, and ground-level imagery into our deep learning models. We will also demo an asynchronous COG tiler (TMS/WMTS/WMS) which will be open sourced prior to the conference and serves map tiles and static images from COGs indexed in a STAC catalog.
“OpenDroneMap: DSMs, DTMs, and Darn Good Orthophotos”
India Johnson;
Talk
OpenDroneMap is a highly customizable, open-source photogrammetry software built, updated, and maintained by a robust contributing community consisting of several parts. OpenDroneMap has emerged as the open-source solution to processing aerial imagery. It is a demonstrated peer of its commercial counterparts, and generates accurate, precise orthophotos, DTMs, and DSMs.
“OpenLayers Feature Frenzy”
Andreas Hocevar, Tim Schaub;
Talk
Ten years ago, OpenLayers was the number one choice for a web mapping library. With the rewrite in 2012 and the arising competition of Leaflet and Mapbox GL JS, there was a phase when the project lost popularity. Today, it has found its niche as a full-featured, flexible and high-performance mapping library that users can count on for the long haul, expecially when their mapping needs get more complex.
“Redistricting with QGIS”
Blake Esselstyn, John Holden, Megan Gall;
Talk
One geospatial task with large implications for many democracies is the drawing of representatives' election districts, or redistricting. This talk will show how people can now harness the strengths of QGIS, using a plugin introduced in 2020, not only to create new redistricting plans, but also to address some of the particular problems and challenges that redistricting presents. We will briefly present an example workflow using the software features as well as actual case studies.
“Seeing through Cloud”
Lydia;
Talk
Being a company that is based in Equatorial Africa, and is using satellite imagery to monitor other parts of the world, its amazing how we can’t get any good imagery for our own backyard. We are building products around satellite imagery and we couldn’t exploit this great data source to our advantage because most of the images for Uganda and most of East Africa is covered in cloud, most of the year. Enter cloud masking tools, SAR and making due with what we can get and drawing conclusions basing on that.
“State of GDAL”
Even Rouault;
Talk
State of GDAL presents the latest updates on the GDAL community and new features, among which GDAL 3.1
“State of GeoServer”
Andrea Aime, Jody Garnett;
Talk
State of GeoServer provides an update on our community and reviews the new and noteworthy features for 2020. GeoServer is a web service for publishing your geospatial data using industry standards for vector, raster and mapping. We have a lot to cover for 2.16 and 2.17 releases, and a preview of the September 2.18 release.
Attend this talk for a cheerful update on what is happening with this popular OSGeo project, whether you are an expert user, a developer, or simply curious what GeoServer can do for you.
“State of GRASS GIS”
Helena Mitasova;
Talk
GRASS GIS is a well established, all-in-one geospatial number cruncher with Python, command line, and GUI interfaces. This talk highlights the latest updates in GRASS GIS 7.8 and reveals plans for version 8.0. Major developments include Python 3 support, transition to PROJ 6, intuitive data management, virtual raster mosaic and improved data compression. The development successfully migrated to GitHub and several Docker images are available. Interfacing with GRASS GIS is now easier with direct batch processing, updated rgrass7 package, QGIS plugin/processing, mapcalc updates and actinia cloud integration with openEO GRASS driver.
“The secret life of open source developers”
Andrea Aime, Ian Turton;
Talk
Ever had a disagreement inside an open source community? Wondered why the developers are not serving the users needs as you’d expect? This talk by two veteran open source developers will help users see the day to day community interaction from our the developer point of view. We’ll look at the reasons that drive people to share their code, the licencing conditions covering it, the real life of developers and associated constraints, and what is actually reasonable to expect from both sides, in order to build a productive and sustainable community.
“Toward a sustainable social and technical software ecosystem to enable geospatial-inspired innovation and discovery”
Shaowen Wang;
Talk
This talk discusses a project funded by the National Science Foundation for engaging diverse communities to conceptualize a national Geospatial Software Institute (GSI) in the United States. The mission of the GSI should be to transform geospatial software, data science, and cyberinfrastructure across many fields to revolutionize diverse discovery and innovation by enhancing computational transparency and reproducibility. GSI is well-positioned to nurture a high-performance, open, and sustainable geospatial software ecosystem across academia, government, and industry.
“Unleash the Hexagons! Combining addresses with H3 and Elasticsearch for insurance exposure analysis”
Tomas Holderness;
Talk
This talk introduces the use of Uber’s Hexagonal Hierarchical Spatial Index (H3) with Elasticsearch to aggregate and query millions of point-locations on-the-fly. In contrast to administrative boundaries, H3 offers a globally uniform polygon-set that works at multiple spatial scales. Using an example from the Insurance sector the talk will demonstrate how an H3 Index can be used with Elasticsearch to filter, aggregate, and map large numbers of points with speed and efficiency. Our application enables insurers to instantly see how many of their locations will be affected by floods, storms and wildfires around the world.
“Using open source software to assess caribou habitat and connectivity over time”
Doug Piercey;
Talk
Caribou populations are at various levels of risk throughout Canada and reasons behind the issue can be complex. A project is underway in Newfoundland and Labrador, Canada to assess the impacts of change on caribou habitat and connectivity over time, including projected changes due to climate change. The goal of this project is to inform decisions related to caribou management and climate change adaptation. The project utilizes open source products for many aspects of the analysis, from the generation of individual animal kernels to the projections of future landscapes to web-based exploration of data.
“What every GIS user should know about projections”
Ian Turton;
Talk
A frequent query raised by new comers to QGIS (and other GIS systems) is "Why
don't my layers line up?". This is usually followed by a long list of things
they have vainly tried to make them line up, randomly changing the canvas and
layer projections with no luck.
At the end of this talk new (and older) users will have more of an understanding
of what projections are and how and why they need to deal with them.
“You don't need a basemap?!”
Mila Frerichs;
Talk
We want everything in our lives, especially online, to be context-aware and display only relevant information.
Our world is full of distractions, and we, as creators of (online) maps, add to it.
We create maps that overload our audience with information and hide essential details in our maps.
People can only retain a limited amount of information, and we should be mindful of that.
In this talk, I will show you techniques on how to reduce distractions in your maps and show lots of examples of distraction-free maps. And answer the talk title question.
Let your data shine.