services/detection_service.py

import torch
import numpy as np

def detect_faults_solar(model, image):
    # Preprocess image
    img_tensor = torch.from_numpy(image).permute(2, 0, 1).float() / 255.0
    img_tensor = img_tensor.unsqueeze(0)

    # Perform inference using ultralytics API
    results = model.predict(img_tensor, verbose=False)

    # Process results
    faults = []
    for detection in results[0].boxes:  # ultralytics API
        class_id = int(detection.cls)
        if class_id == 0:
            faults.append({"type": "Crack", "location": (detection.xyxy[0][0].item(), detection.xyxy[0][1].item())})
        elif class_id == 1:
            faults.append({"type": "Damage", "location": (detection.xyxy[0][0].item(), detection.xyxy[0][1].item())})
        elif class_id == 4:
            faults.append({"type": "Hotspot", "location": (detection.xyxy[0][0].item(), detection.xyxy[0][1].item())})

    return faults

def detect_faults_windmill(model, image):
    # Preprocess image
    img_tensor = torch.from_numpy(image).permute(2, 0, 1).float() / 255.0
    img_tensor = img_tensor.unsqueeze(0)

    # Perform inference using ultralytics API
    results = model.predict(img_tensor, verbose=False)

    # Process results
    faults = []
    for detection in results[0].boxes:
        class_id = int(detection.cls)
        if class_id == 2:
            faults.append({"type": "Blade Damage", "location": (detection.xyxy[0][0].item(), detection.xyxy[0][1].item())})
        elif class_id == 3:
            faults.append({"type": "Motor Fault", "location": (detection.xyxy[0][0].item(), detection.xyxy[0][1].item())})

    return faults