Good Morning! Colleagues, can anyone suggest solutions for classifying different categories using multispectral images from natural resource satellites.
Hi there @rudiney,
not sure so asking if data you already shared in this topic is related to current request or those are two separate issues?
Br,
Ivan
Hi there @ipazin,
The data are related. My goal is to carry out an analysis on the data trying to identify the possible relationships between the spectral behavior of the vegetation and the types of soil treatment
You may find this example helpful:
The data isn’t satellite data, but microscope images. It uses deep learning to classify the images as different types of cancer. You could try to use a similar strategy to analyze your soil types.
Here’s a related blog post:
https://www.knime.com/blog/transfer-learning-made-easy-with-deep-learning-keras-integration
4 Likes
Hi @rudiney,
Could you elaborate a bit more on your input images/data and even provide sample data?
One aspect that is particularly interesting w.r.t. @Corey’s workflow: Do you have one soil treatment associated with one (multispectral) image?
Best,
Stefan
Hi @rudiney! I am attempting to do something similar to identify tillage practices. I am working on constructing a CNN at the moment.
Were you able to carry out the analysis that you wanted to? Which satellite images did you use?
This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.