0

I have created and trained an object detection model in IBM Watson studio to detect a specific part of images. I was wondering how would I then extract that detected image and store it as a separate image? Is there something built into IBM Watson that does this or would I have to use something else to integrate the model with? If I do have to use something else, what would that be and how would I do it?

Thank you in advance!

dontknow
  • 1

1 Answers1

0

IBM Watson doesn't have anything to do that extraction that I'm aware of. But a quick search showed a few utilities that can do that. The API returns the location of the detected object as a bounding box with X and Y positions and width and height.

Allen Dean
  • 597
  • 2
  • 10
  • Thank you! As a follow-up, What application or software that is compatible with IBM Watson Studio can I use that is able to utilize these coordinates to extract the image? – dontknow Jul 14 '20 at 18:51
  • I don't know that there's anything built into Watson Studio to support this. I think that you'll need to get the details from by using the API. However, take a look at help topic at https://dataplatform.cloud.ibm.com/docs/content/wsj/analyze-data/visual-recognition-development.html – Allen Dean Jul 15 '20 at 11:47