Upload app.py

app.py

@@ -0,0 +1,61 @@
+import gradio as gr
+import numpy as np
+import keras
+import torch
+from PIL import Image
+from keras.preprocessing import image as keras_image
+
+# load the model
+def load_model(model_path):
+    model = keras.models.load_model(model_path)
+    return model
+
+
+# Function to preprocess the input image
+def preprocess_image(image):
+    # Convert image to grayscale
+    image = np.array(image)
+    image = Image.fromarray(image).convert('L')
+    
+    # Resize the image
+    image = image.resize((128, 128))
+    
+    # Convert to numpy array and normalize
+    image = np.array(image)
+    image = image / 255.0
+    
+    # Add batch dimension
+    image = np.expand_dims(image, axis=-1)
+    
+    # Stack the grayscale image to make it a 2-channel image
+    image = np.repeat(image, 2, axis=-1)
+    
+    # Add batch dimension
+    image = np.expand_dims(image, axis=0)
+    
+    return image
+
+# Function to perform segmentation prediction
+def predict_segmentation(model, image):
+    segmentation_map = model.predict(image)[0]
+    threshold = 0.5
+    segmented_image = (segmentation_map > threshold).astype(np.uint8)
+    segmented_image = Image.fromarray(segmented_image * 255)
+    return segmented_image
+
+
+# Define Gradio interface
+def gradio_interface(model_path):
+    # Load the model
+    model = load_model(model_path)
+
+    # Define input and output components
+    demo = gr.Interface(lambda image: predict_segmentation(model, preprocess_image(image)),
+                       inputs = "image" ,
+                       outputs= "image" ,
+                       title = "Brain Tumor Segmentation",
+                       description = "Upload an image of a brain scan, and the model will segment the brain tumor.")
+ 
+    demo.launch(share=True)
+
+gradio_interface("model_x81_dcs65.h5")