patelmmayur003/solar-snow-detection-v1

Model Card: Solar Snow Segmentation v1

This model performs image segmentation to classify snow accumulation on solar panels.
It is trained using Ultralytics YOLO segmentation and helps solar operators identify:

• Clean panels
• Partially snow-covered panels
• Fully snow-covered panels

This supports automated O&M, energy yield prediction, and maintenance planning.

---

Model Details

Model Description

This model is a YOLO-based segmentation model trained on real solar panel images captured during winter conditions.
The goal is to automatically segment the snow regions on PV modules and quantify their impact.

• Developed by: Mayur Patel
• Model type: YOLO segmentation model (Ultralytics)
• Task: Image segmentation
• Classes:
  • 0: Clean_Panel
  • 1: Snow_Partial
  • 2: Snow_Covered
• License: MIT
• Finetuned from model: YOLOv8n-seg (Ultralytics)

Model Sources

• Repository: https://huggingface.co/patelmmayur003/solar-snow-detection-v1
• Dataset: Custom dataset (Snow_Loss_Trail_08)
• Demo: Coming soon (Gradio app optional)

---

Uses

Direct Use

Users can load the model with Ultralytics:

from ultralytics import YOLO
model = YOLO("patelmmayur003/solar-snow-detection-v1")
results = model("test.jpg")
results.show()