Road Sign Extraction with Deep Learning Example?

Question

Hi all. I attended the imagery summit yesterday and saw the example of extracting road signs from video using deep learning again. We will be doing a road sign inventory soon and would like to try this workflow. I have seen this showcased several times now in different ESRI presentations\forums. I am wondering if there are any resources for this workflow, particularly any training datasets and recommendations for equipment and capture specs? The closest thing I could find is Automate Road Surface Investigation Using Deep Learning example which I'm sure is similar. Thanks again.

LukeHenderson1 · Answer

Is there anything new with this? Gathering GIS data for DOT assets from cameras would be beneficial project.

MehdiPira1 · Answer

Hi Devin Doring‌,

I've found the following feature services for Stop Signs:

https://www.arcgis.com/home/item.html?id=f4e7ad6b7c6e43758e22d910f1938c40

https://www.arcgis.com/home/item.html?id=459dc422538c478b8565098027333c10

danretired · Answer

there is a data folder but not anything specific to road signs, that is just where the notebooks and the data are housed

GitHub - Esri/arcgis-python-api: Documentation and samples for ArcGIS API for Python

check inside the Samples folder for other bits