Pylon Detection - USA

A pylon is an essential structure in various industries. It serves roles like supporting power lines or signaling important locations. Detecting pylons in high-resolution satellite or aerial imagery is crucial for infrastructure maintenance and ensuring power grid reliability. Detecting pylons is important for safety, as it can help in identifying damaged or fallen pylons. It can also prevent accidents by ensuring clear visibility of road signs and signals. Detecting pylons can be useful for optimizing traffic flow and aiding in urban planning decisions. Use this deep learning model to automate the task of detecting pylons in high-resolution satellite or aerial imagery.

Using the model
Follow the guide to use the model. Before using this model, ensure that the supported deep learning libraries are installed. For more details, check Deep Learning Libraries Installer for ArcGIS.

Fine-tuning the model
This model can be fine-tuned using the Train Deep Learning Model tool. Follow the guide to fine-tune this model.

Input
8-bit, 3-band high resolution (60 centimeters) imagery. 

Output
A feature class representing detected pylons. 

Applicable geographies
The model is expected to work well in the United States.

Model architecture
This model uses the MMDetection based Dynamic-RCNN model architecture implemented in ArcGIS API for Python.

Accuracy metrics
This model has an average precision score of 0.95.

Training data
This model has been trained on an Esri proprietary pylon detection dataset.

Limitations
The model does not work well in urban areas as it gets confused by the sharp shadow of the trees.

Sample results
Here are a few results from the model.