Detect Objects Using Deep Learning error

MichaelMichaelides — Wed, 16 Sep 2020 10:23:04 GMT

Hello. I am a newbie about deep learning and I was trying to run the "Detect Objects Using Deep Learning" tool to detect buildings in a tiff aerial image an I got an error (see below). I'm using ArcGis Pro 2.6.1.

I followed the workflow successfully, meaning that I created the labeled objects, trained the model - no errors at these stages. When I tried to run the "Detect Objects Using Deep Learning" tool I got the following error message:

"ERROR 999999: Something unexpected caused the tool to fail. Contact Esri Technical Support (http://esriurl.com/support) to Report a Bug, and refer to the error help for potential solutions or workarounds.
Unable to prepare input raster(s) for the python raster function.
The specified non-optional parameter doesn't have a value. [rasters]
Failed to execute (DetectObjectsUsingDeepLearning)."

For the export of training data i used the Meta data format: Labeled Tiles. Note that when I first ran the export i used the Pascal Visual Object Classes and i had no problem at all. I mean the tool of detection ran successfully but the results were not so good so i decided to change the Meta data format to Labeled Tiles and i saw that the Image chips were better than the first time but the problem as i mentioned was not able to run the last tool (detecting objects)

In the following image are the parameters for the training model.

This the error message with the parameters used.

Can you help me please?

Mike#

Re: Detect Objects Using Deep Learning error

TAN860421 — Wed, 27 Jan 2021 07:22:35 GMT

I also encountered the same problem. I am currently testing the deep learning FasterRCNN and also encountered a warning problem. Even if I refer to the relevant information on the Internet, it still has not improved. Do you have any solutions? Can you help, please?

topic Detect Objects Using Deep Learning error in ArcGIS Pro Questions

Detect Objects Using Deep Learning error

Re: Detect Objects Using Deep Learning error