Data og algoritmer skal hjælpe med at vedligeholde vores kloakker

Press/Media: Press / Media

Description

Netop i disse dage offentliggøres unikt datasæt med 1,3 millioner billeder fra danske kloakker. Det skal gøre vedligehold af rørene nemmere.

Period	21 Jun 2021

Media contributions

1

Media contributions

Title	Data og algoritmer skal hjælpe med at vedligeholde vores kloakker
Degree of recognition	National
Media name/outlet	Videnskab.dk / ForskerZonen
Media type	Web
Country/Territory	Denmark
Date	21/06/2021
URL	https://videnskab.dk/forskerzonen/teknologi-innovation/data-og-algoritmer-skal-hjaelpe-med-at-vedligeholde-vores-kloakker
Persons	Joakim Bruslund Haurum, Thomas B. Moeslund

Keywords

Sewer Inspection
Sewer
Computer Vision
Defect classification
Dataset

Related content

Publications
Sewer-ML: A Multi-Label Sewer Defect Classification Dataset and Benchmark
Research output: Contribution to book/anthology/report/conference proceeding › Article in proceeding › Research › peer-review
A Survey on Image-Based Automation of CCTV and SSET Sewer Inspections
Research output: Contribution to journal › Review article › peer-review

Press/Media
NVIDIA DGX-2 AI System Plays Crucial Role in Sewer Research
Press/Media: Press / Media
Unikt datasæt skal bidrage til vedligehold af danske kloakker
Press/Media: Press / Media
Supercomputer spiller afgørende rolle i kloak-forskning
Press/Media: Press / Media

Impacts
Robots find defects in sewers
Impact: Economic impact, Other impact

Projects
ASIR: Autonomous Sewer Inspection Robot
Project: Research

Datasets
Sewer-ML
Dataset