nav[aria-label="Primary Navigation"] { padding: 0; & ul { list-style: none; width: 100%; display: flex; flex-direction: row; justify-content: start; align-items: start; gap: 30px; padding: 0; & li { margin: 0; } & ul li { list-style: none; } } }

How can I find out more about the Feature Classifier model type for Object Classification using Deep Learning?

ScarletMaguire

I will be building some Object Classification models in ArcGIS Pro. The only model type available for this is the "Feature Classifier" model type. However, there seems to be little literature from ESRI about this model type and how it works. Can anyone recommend any documentation, published by ESRI or someone else?

Find more posts tagged with

Deep learning

Accepted answers

All comments

PavanYadav

@ScarletMaguire please see if these help:

How feature classifier works?

Coastline classification using Feature Classifier

ScarletMaguire

Hi Shivani. Thanks for your reply. If by cell size you mean resolution, the cell size is about 10cm. I collected this data with a UAV. I do actually have a polygon feature class which outlines the objects that need to be classified (I forgot to mention this in my original post).

The thing is- I do not need help in training the model itself, I have already experimented with the Deep Learning workflow in ArcGIS Pro and it performs well for what I need it to do. What I need help with is understanding the actual Deep Learning algorithm/architecture that is being used when I choose the "Feature Classifier" model type. To put this into context, when I build an Object Detection model, for example, by using Mask R-CNN, this is a defined deep learning model architecture that I can go and read about. However, I simply cannot find any literature regarding an architecture called "Feature Classifier". This is what I need help with.

ShivaniPathak

Hi @ScarletMaguire for training an object classification model you need an input raster and a feature class which represents the classes. Can you share details about the spatial resolution (cell size) and sensor of your input raster?

ScarletMaguire

I am looking to classify damaged trees into different categories depending on how damaged they are. My source data is raster. Thanks for the link to the help articles.

ShivaniPathak

Hi @ScarletMaguire if you want to train an object classification model you can refer to these sample notebooks:

You can refer to the guide for technical concepts of Feature Classifier: https://developers.arcgis.com/python/latest/guide/how-feature-categorization-works/

danretired

what objects are you looking to classify and is the source data vector or raster?

There is a whole section in the help files

https://pro.arcgis.com/en/pro-app/latest/tool-reference/image-analyst/classify-objects-using-deep-learning.htm

other resources exist, but I suspect you may need to clarify or narrow down possible topic links

Join the Conversation

Signed in members can post, follow updates, and more. New here? Register a free account.

Quick Links

Community Help
Find useful guides, FAQs, and documents to help you navigate and make the most of Esri Community.