OSM Kenya/Projects/Open Hardware GNSS and Its Applications in OpenStreetMap Imagery Alignment
Introduction
Imagery offsets in OpenStreetMap (OSM) refer to the discrepancy between the position of features depicted in aerial or satellite images and their actual position on the Earth's surface. These offsets can be caused by a range of factors, including the perspective of the imagery, the Earth's curvature, and the precision of the images themselves. To maintain the accuracy and utility of the map data within OSM, it is essential to rectify these offsets by aligning the imagery with the corresponding features on the ground. By addressing imagery offsets through rectification, mappers can contribute to the reliability and currency of OSM as a resource for individuals worldwide.
In OpenStreetMap (OSM), precise GNSS points refer to locations on the Earth's surface that have been accurately surveyed and recorded using Global Navigation Satellite System (GNSS) technology. The preferred tag on OSM is man made=survey point. These points can be used as reference points for aligning aerial or satellite imagery, as well as for verifying the accuracy of other map features. Precise GNSS points can be collected using a variety of GNSS receivers, such as handheld devices or vehicles equipped with specialized sensors(Mapillary comes to mind). The accuracy of these points can be highly precise, often within centimeters or even millimeters. By incorporating precise GNSS points into OSM, mappers can help ensure the overall accuracy and quality of the map data.
There can be a shortage of precise physical survey points in OpenStreetMap (OSM) in some geographical regions such as Africa. A quick search on Overpass Turbo can verify this (most of the points are sourced from USGS through remote sensing). This could be due to a lack of surveying or mapping activity in that region, or it could be because the terrain or environment makes it difficult to collect accurate GNSS data. Another possibility is that there may be a shortage of mappers or other volunteers available to contribute GNSS points to OSM. This could be due to a lack of awareness about the importance of these points, or it could be because there are not enough resources or incentives in place to encourage mapping activity. Regardless of the cause, a shortage of precise GNSS points in OSM can impact the overall accuracy and quality of the map data.
Background
Open hardware refers to physical objects or devices that are designed and shared in a way that allows anyone to study, modify, distribute, make, and sell the hardware based on that design. This approach is in contrast to traditional proprietary hardware, which is typically owned by a company and protected by intellectual property laws. Open hardware can take many forms, such as electronic devices, machines, and even buildings. It is often associated with the principles of open source, which emphasize collaboration, transparency, and the sharing of knowledge. By embracing open hardware, individuals and organizations can promote innovation, foster community, and encourage the creation of new products and solutions.
Local OpenStreetMap (OSM) communities are groups of individuals who are interested in contributing to and promoting OSM in a specific geographic region. These communities can take many forms, such as online forums, meetup groups, or local chapters of national or international organizations. Local OSM communities are important for a number of reasons:
- They provide a way for mappers in a particular area to connect with each other and share information and resources.
- They can help coordinate mapping efforts and encourage collaboration among mappers.
- They can help raise awareness about OSM and its potential uses in the local community.
- They can provide support and training for new mappers, helping to grow the OSM community.
Overall, local OSM communities can play a vital role in supporting and advancing the goals of OSM and can help ensure that the map data is accurate and up-to-date for a particular region.
OSM Kenya
OSM Kenya aims to create a vibrant community of OpenStreetMap contributors and users in Kenya and provide a platform where members can exchange ideas and support each other. In the years to come,we hope to fully map the country on OpenStreetMap and to strongly advocate for its use withindifferent groups such as government, universities, and organizations. This can only be done sustainably by mapping while applying OSM in creating solutions that we face locally. By working on similar projects,we will create an open and diverse community of OpenStreetMap contributors in Kenya, develop and support local initiatives that are aimed at improving OpenStreetMap in Kenya, and promote the use of open data, including OpenStreetMap and FOSS countrywide. We hope to engage with different organizations, communities, and government institutions as we continue to grow.
Methodology
- The area to be mapped was identified as a Clay City ward in Kasarani SubCounty of Nairobi.
- The resources needed for the mapping project were determined, including two sets of GNSS receivers (one being an innovation of low-cost open hardware and the other, a commercial-grade GNSS system), connection to a CORS network from the Regional Centre for Mapping, and Resources for Development - RCMRD, JOSM, and a car for transport within the area.
- Volunteers from OSM Kenya were recruited to help with the mapping project after several meetings and email correspondence.
- A date and time for the mapping event was set, ensuring good weather and sufficient volunteer availability.
- A training session was prepared for volunteers who were new to GNSS mapping, covering the basics of using the GNSS receivers and mapping software, as well as any safety considerations.
- On the day of the mapping event, the volunteers gathered, and the GNSS receivers were set up.
- A plan for the mapping route was set out, ensuring that all desired areas were covered. This was discussed prior so as to ensure that the survey points to be captured were already existing points that were visible on the current imagery in OSM. Also, this was done as a point to minimize our budget and avoid setting up survey points in a more traditional land survey way.
- The volunteers walked and drove the mapping route, using the GNSS receivers to capture precise location data for OSM. Rather than collect the points using Static GNSS surveying, we decided to conduct a survey using rovers connected to a CORS network while verifying with a base station that was set up in the neighborhood (< 4 km). In future works, we could look at doing Static surveys and post-process the data through precise point positioning - PPP.
Once the data was collected, it was extracted in CSV format and the raw .ubx and .rinex formats were stored. Since we had a commercial-grade GNSS, we did a double collection of all the points as a sort of accuracy check for the open hardware instrument. After combining the data and getting satisfied with the accuracy and precision of the data, we uploaded it to OSM using JOSM. The data was reviewed and any necessary edits or corrections were made. See the photos
Results
A search on Overpass turbo or OSMCha can be used to view the uploaded changesets. These points can be applied in aligning imagery on JOSM. For this particular region, the features on OSM are best aligned with Bing satellite imagery and so it is easier to align any other imagery with ground-truth data.
| 266741.071 | 9865836.998 |
|---|---|
| 266726.105 | 9865892.766 |
| 264823.902 | 9864517.812 |
| 264826.956 | 9864480.539 |
| 267238.619 | 9864640.796 |
| 267590.984 | 9865360.932 |
| 267485.683 | 9865721.039 |
| 266741.0943 | 9865836.986 |
| 264823.9382 | 9864517.735 |
Discussion
The low-cost hardware was great in accuracy when it had an RTK Fix. However, in areas with canopies of different sorts, for example, buildings and trees, it was more difficult to obtain an RTK Fix. On this front, it was easier for the commercial-grade instrument. This is easily attributed to a bottleneck in the antenna choice rather than the GNSS processor itself. An obvious advantage to using open hardware is the ability to attach a better antenna according to the use case. For instance, drone applications can use helical antennas, while survey work can use choke-ring antennas.
We countered the challenge of getting RTK fixes using the commercial-grade GNSS.
In terms of the number of instruments needed, it is much more cost-efficient for local OSM communities to build, buy, or rent low-cost GNSS equipment. Financial resources can then be spent on volunteers or other expenses. To put it in perspective, a fully-fledged commercial GNSS can cost $10000 while low-cost equipment can cost as little as $400.
The logistics of the work: a vehicle is required to move personnel and equipment. If a known control point is used as a base station, a person is needed to watch over the instrument for purposes of ensuring no movements in the XYZ axes but also for security.
Considering that the data collection is based on fieldwork, it is necessary to ensure that volunteers are remunerated. Therefore, budgeting for a project of a similar nature should include the costs of hiring volunteers, hiring, buying, or building a GNSS instrument, hiring a vehicle, security personnel if necessary, and other miscellaneous costs such as insurance and permits.
Conclusion
The concepts of imagery offsets, open hardware, and precise GNSS points are all integral to the OpenStreetMap (OSM) landscape. Imagery offsets denote the discrepancy between the position of features depicted in aerial or satellite images and their actual position on the Earth's surface. Rectifying these offsets is crucial for maintaining the accuracy and utility of the map data within OSM. Local OSM communities are groups of individuals who are interested in contributing to and promoting OSM in a specific geographic region. These communities can provide support and training for new mappers, facilitate coordination of mapping efforts, and increase awareness about OSM in the local community. OSM chapters are local or regional organizations that are affiliated with the OpenStreetMap Foundation (OSMF) and are established to support and advance the goals of OSM within a specific geographic region or community. The methodology for capturing precise GNSS points for OSM through local OSM communities involves identifying the area to be mapped, determining the resources required, recruiting volunteers, setting a date and time for the mapping event, preparing a training session for new mappers, gathering the volunteers and equipment, establishing a mapping route, capturing the GNSS data, uploading the data to OSM, and reviewing and making any necessary edits or corrections. By adhering to this methodology, local OSM communities can successfully collect precise GNSS points for OSM and enhance the accuracy and quality of the map data.