Segment Burns API

SkinBurns_Segmentation.py

@@ -0,0 +1,68 @@
+import numpy as np
+from skimage import io, img_as_float, color, transform
+from scipy.ndimage import gaussian_filter
+import tensorly as tl
+from tensorly.decomposition import tucker
+from sklearn.cluster import KMeans
+import cv2
+import matplotlib.pyplot as plt
+
+def segment_skin_burns(image_path, output_path='segmented_image.png'):
+    # Load and preprocess the image
+    image = io.imread(image_path)
+    image = img_as_float(image)
+    image_lab = color.rgb2lab(image)  # LAB color space
+
+    # Adaptive Illumination Correction (CLAHE)
+    clahe = cv2.createCLAHE(clipLimit=3.0, tileGridSize=(8, 8))
+    image_lab[:, :, 0] = clahe.apply((image_lab[:, :, 0] * 255).astype(np.uint8)) / 255.0
+
+    # Gaussian filtering
+    sigma_value = 1.5
+    image_lab_filtered = np.zeros_like(image_lab)
+    for i in range(3):
+        image_lab_filtered[:, :, i] = gaussian_filter(image_lab[:, :, i], sigma=sigma_value)
+
+    # Third-order tensor
+    tensor_image = tl.tensor(image_lab_filtered)
+
+    # Tucker3 tensor decomposition
+    tucker_rank = [30, 30, 3]
+    core, factors = tucker(tensor_image, rank=tucker_rank)
+
+    # Resize core tensor
+    core_resized = transform.resize(core, (image.shape[0], image.shape[1], 3), anti_aliasing=True)
+
+    # Convert to RGB
+    core_resized_rgb = np.clip(color.lab2rgb(core_resized), 0, 1)
+
+    # Reconstruct image
+    reconstructed_image = tl.tucker_to_tensor((core, factors))
+    
+    # Features for clustering
+    height, width, _ = image.shape
+    feature_vectors = reconstructed_image.reshape(height * width, -1)
+
+    # K-means clustering
+    n_clusters = 3  # Burn wound, healthy skin, background
+    kmeans = KMeans(n_clusters=n_clusters, n_init=10, random_state=42)
+    kmeans.fit(feature_vectors)
+    labels = kmeans.labels_.reshape(height, width)
+
+    # Identify burn cluster
+    cluster_means = [np.mean(feature_vectors[labels.flatten() == i], axis=0) for i in range(n_clusters)]
+    burn_cluster = np.argmax([mean[1] for mean in cluster_means])  # LAB 'a' channel: higher values → red tones
+
+    # Color-mapped segmentation image
+    segmentation_map = np.zeros((height, width, 3))
+    colors = {burn_cluster: [1, 0, 0],  # Red for burn
+              (burn_cluster + 1) % 3: [0, 1, 0],  # Green for healthy skin
+              (burn_cluster + 2) % 3: [0, 0, 1]}  # Blue for background
+
+    for cluster, color in colors.items():
+        segmentation_map[labels == cluster] = color
+
+    # Save the segmented image
+    plt.imsave(output_path, segmentation_map)
+
+    return output_path

app.py

@@ -4,6 +4,7 @@ from pymongo.server_api import ServerApi
 import cloudinary
 import cloudinary.uploader
 from cloudinary.utils import cloudinary_url
+from SkinBurns_Segmentation import segment_skin_burns 
 
 app = FastAPI()
 
@@ -57,6 +58,37 @@ async def upload_sample(file: UploadFile = File(...)):
     except Exception as e:
         return {"error": str(e)}
 
+@app.post("/skin_burns/segment")
+async def Segment_Burns(file: UploadFile = File(...), output_path='segmented_image.png'):
+    try:
+        # Save the uploaded file temporarily
+        with open(file.filename, "wb") as temp_file:
+            temp_file.write(await file.read())
+
+        # Call the segmentation function
+        segmented_image_path = segment_skin_burns(file.filename, output_path)
+
+        # Upload the segmented image to Cloudinary
+        result = cloudinary.uploader.upload(segmented_image_path, public_id=f"segmented_{file.filename}")
+        uploaded_url = result["secure_url"]
+
+        # Save segmented image URL to MongoDB
+        collection = db["Images"]
+        doc = {"filename": f"segmented_{file.filename}", "url": uploaded_url}
+        collection.insert_one(doc)
+
+        return {"segmented_image_url": uploaded_url}
+    except Exception as e:
+        return {"error": str(e)}
+    
+
+
+
+
+
+
+
+
 # ✅ Optimize and transform image URL
 @app.get("/cloudinary/transform")
 def transform_image():

requirements.txt

@@ -2,4 +2,11 @@ fastapi
 uvicorn[standard]
 pymongo[srv]
 python-multipart
-cloudinary
+cloudinary
+numpy
+scikit-image
+scipy
+tensorly
+scikit-learn
+opencv-python
+matplotlib