This model generates annotated images to view and download as well as a spreadsheet that lists the equipment detected, the model’s confidence level for each detection, the total number of equipment found, and more.
The Power Pole Equipment Detection model analyzes utility pole images, classifying the equipment detected on the pole. Model results are an excellent source for asset inventory and training purposes.
Note: Current mAP (Mean Average Precision) score is 63%. Results will continue to be refined as the model is trained on more data.
This model detects a variety of equipment that is typically found on power poles, including:
PNG files
JPEG files
Annotated images
A summary report in XLSX format