Info Page

Step by step instruction

1. Choose data sources and how to draw on map.
   • You start with choosing you data sources for labels and orthography. You can also instead upload a file with metadata instead.
   • Then you choose how you want to choose area on a map
2. Designate area to be used for training.
   • You can draw a square or a polygon on a map,
   • or you can write in the coordinates,
   • or you can upload a geoJSON-file of the area you want to use.
3. On constraints, you fill in the parameters and settings for you training data.
   • Data Settings: Here you enter how much of the data will be used for training, how much will be used for validation and the minimum percentage of buildings. If you are unsure about which values to put in, then set them at 60% for Training, 40% for Validation and 10% for Building percentage.
   • WMS Layers: If you have chosen WMS as label source, you can choose if you want buildings, roads or bridges in the training data and you can choose which colour they will have. This is not possible for NGIS or FGB as label source.
   • Image settings: Here you can define the size of the tiles in pixels of the training data. It must be a value between 100-1104. For image resolution you can choose a value between 0.1-0.5. 0.1 is 10cm.
   • Dataset settings: Here you can choose if you want the data sent to your email and give you dataset a name. If you do not enter your e-mail, you can download the dataset directly to your computer locally.
   • When you have filled all the parameters and required fields, you can order you data by clicking on the order button and then confirm.
   • If you have not entered an email, you will after having clicked order be given the option to download the data.

What can you use the generated data for?

The generated data can be used for training machine learning models.

About KartAi

• A research project for improving the quality of Norway's official property register (cadastre) and central common map database (SFKB) using artificial intelligence (AI).
• The aim is to contribute to automating and streamlining the processing of construction building applications.
• A collaboration between Kristiansand municipality (project owner), Kartverket, Norkart AS and the University of Agder.
• Financially supported by Regional research fund Agder.
• The initiators are Alexander Salveson Nossum, Norkart AS; Terje Nuland, Kristiansand municipality; Lars Fredrik Gyland and Ivar Oveland, Kartverket.

KartAi is a research project conducted by Kartverket, Norkart, Kristiansand Municipality, and the University of Agder with financial support from the Regional Research Fund Agder. The main purpose of the project is to make the processing of building cases in the municipality more efficient. This will be achieved by improving the quality of the property register (matrikkelen), and map databases using Ai. The KartAi project has three sub-goals:

1. Buildings will be identified by the help of different data sources like, laserdata, aerial photo and crowd sourcing. This will be done by creating a “geodata lake”, with data sources that are optimized for Ai and big data analysis.
2. Provide a better decision-making basis for municipal case management. The project will identify buildings under 50 square meters in a selected area in Kristiansand by using and developing new Ai algorithms from the mentioned “geodata lake” This will typically include buildings that are exempt from the municipal application requirements.
3. Ensure more efficient processing of cases through reliable and automated quality improvement of the national property register (matrikkelen) and map databases in dialog with residents.

This application belongs to the first subgoal, the generated data can be used to train Ai to recognize buildings.

Who made this application

This application was developed by two different student groups at the University of Agder

Spring 2023

• Ine Mari Bredesen
• Tobias Ekholt
• Niklas Fugledal
• Kasper Johan Idland Skjæveland
• Sondre Slettene

Spring 2024

• Cassandra Vårdal
• Tørres Båshus
• Gustav Svartsund
• Larisa Saitova
• Victor Bakken

Privacy Policy

Privacy Policy

Last updated: 11.04.2024

This privacy policy contains information about how KartAi Training Data Generator handles your personal data. We encourage you to read the privacy policy carefully.

1. Contact information for KartAi

2. Central concepts

3. KartAi is the data controller

4. What kind of information do we collect?

5. Alternative to email

6. The purpose of the processing of the personal data?

7. How long is the information stored?

8. With whom can the information be shared with?

9. Information sharing with third parties

10. Where is the information stored?

11. Information cookies

1. Contact information for KartAi

KartAi is responsible for handling your personal data when using this application. If you have questions about the processing or wish to get in touch with us to exercise your rights, you will find our contact information below:

KartAi - https://kartai.no/
Contact person: alexander.nossum@norkart.no

2. Central concepts

Personal data is information that alone or together with other information can be used to identify, locate or contact a person. Examples of personal data are name, telephone number and IP address.

Processing of personal data involves all forms of handling of personal data such as: collection, analysis, registration and storage.

The controller is the person who determines the purpose of the processing of personal data and which information is requested. It is the data controller who is responsible for the handling of your personal data in accordance with the applicable Personal Data Act.

3. KartAi is the data controller

KartAi is responsible for handling your personal data.

4. What kind of information do we collect?

When you order training data from the application, we need your email address to be able to send it to you. For that reason, we ask you for your e-mail address. The email address is not saved.

5. Alternative to mail

An alternative to email is to download training data directly from the site.

6. The purpose of the processing of the personal data?

We need your email address in order to send training data to you instead of downloading directly from the site.

7. How long is the information stored?

We only process the personal data for as long as it takes to fulfill the purpose of their collection, after which we delete the information. We do not store the e-mail address or any other personal data.

8. With whom can the information be shared?

The information is not shared with anyone.

9. Information sharing with third parties

We do not sell your personal data to third parties.

10. Where is the information stored?

No personal data is stored.

11. Cookies

Cookies are small text files that are placed on your computer when you download a website.

Storage of information and processing of this information is not permitted unless the user has both been informed about and has given his consent to the processing. The user must find out about and approve which information is processed, what the purpose of the processing is and who processes the information.

We only use cookies to give you a better user experience on our pages so that you don't have to fill in fields again every time you are on our pages.

Privacy statement - Norwegian Data Protection Authority
Read what the Norwegian Data Protection Authority says about privacy statements: https://www.datatilsynet.no/rettigheter-og-plikter/virksomhetenes-plikter/informasjon-og-apenhet/

Close

Cookie Policy

Cookie Policy for Training Data Generator

Last updated: 18.03.2024

At KartAI, we believe in being clear and open about how we collect and use data related to you. This policy provides detailed information about how and when we use cookies. By using Training Data Generator, you agree to our use of cookies as described in this Cookie Policy.

What is a cookie?

A cookie is a small file placed onto your device that enables Training Data Generator features and functionality. For example, cookies enable us to identify your device and secure your access to Training Data Generator.

Why does Training Data Generator use cookies?

Cookies play an important role. They help us provide you with a seamless experience by maintaining session continuity and safe connections as you navigate through the application. Here's a brief overview of the types of cookies we use:

• Session Cookies: These are temporary cookies used to remember you during the course of your visit to the application, and they expire when you close the web browser.

How does Training Data Generator use cookies?

Sessions Management: Cookies help us keep track of your activity as you navigate through the application, ensuring that the application remembers your activity.

What to do if you don't want cookies to be set

You can usually manage cookies through your browser settings. Most browsers provide the ability to block cookies or block cookies from specific sites. Browsers can also help you delete cookies when you close the application. However, you should note that our application requires cookies to function properly, and disabling cookies may affect your experience.

Changes to this policy

We may update this policy from time to time. Please revisit this page to stay aware of any changes. By continuing to use Training Data Generator and our services after those changes become effective, you agree to be bound by the revised Cookie Policy.

Close