Skip to main content
Question

Computer vision - recognise feature type in photo and calculate area coverage?

  • May 15, 2024
  • 1 reply
  • 80 views

p_c_20
Contributor
Forum|alt.badge.img+8

I’m looking at a workflow wherein the start point is this: 

  • Underwater photo manually copied into Powerpoint
  • Measurement grid overlaid
  • Percentage cover of different plant species estimated by eye
  • Results copied to Excel

I haven’t dabbled in CV at all within FME, just read this blog - and this may be jumping in the deep end - but am wondering if this scenario might be a candidate? 

1 reply

donalmateer
Safer
Forum|alt.badge.img+5

Hey @p_c_20 

Thank you for the question!

I believe FME has a few options when it comes to Computer Vision which may help you get to the answer.

This blog post may be a good starting point. (https://fme.safe.com/blog/2020/06/build-computer-vision-workflows-picterra-fme/). Picterra is a separate platform which we integrate into FME.

 

There is also ways to detect features on photos with standard FME tools such as:
RasterExpressionEvaluatorRasterConvolver and maybe point cloud transformers. 


FME is not a real feature detection/feature extraction tool, but in some situations, for certain types of imagery, it might work. 
If you need further assistance would you possibly share some sample data and we could look into finding a solution.


Cookie policy

We use cookies to enhance and personalize your experience. If you accept you agree to our full cookie policy. Learn more about our cookies.

 
Cookie settings