Upload 9 files

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+fowl_pox_model.keras filter=lfs diff=lfs merge=lfs -text

app.py

@@ -0,0 +1,75 @@
+from keras.models import load_model
+import cv2
+import gradio as gr
+import os
+
+pox_model = load_model('fowl_pox_model.keras', compile=True)
+class_name = {0: 'Healthy', 1: 'Chicken have fowl pox', 2: 'Unknown'}
+status = {0: 'Non Critical', 1: 'Critical', 2: 'N/A'}
+recommend = {0: 'No need medicine', 1: 'Panadol', 2: 'N/A'}
+
+def predict(img):
+    # Resize the image to the required size for the model
+    img_resized = cv2.resize(img, (256, 256))
+    
+    # Make the prediction
+    pred = pox_model.predict(img_resized.reshape(1, 256, 256, 3)).argmax()
+    
+    # Get the prediction details
+    prediction_label = class_name[pred]
+    prediction_status = status[pred]
+    recommendation = recommend[pred]
+    
+    # If the predicted label is 1, draw the bounding boxes on the image
+    if pred == 1:
+        # Example: Assuming the model provides bounding box coordinates in a list of tuples
+        # Each tuple is (x, y, w, h) for the bounding box
+        bounding_boxes = [(50, 50, 100, 100), (150, 150, 50, 50)]  # Example coordinates
+        
+        # Create a copy of the original image
+        img_with_boxes = img.copy()
+        
+        # Scale factor to adjust bounding box coordinates
+        scale_x = img.shape[1] / 256
+        scale_y = img.shape[0] / 256
+        
+        # Calculate the merged bounding box
+        min_x = min([x for (x, y, w, h) in bounding_boxes])
+        min_y = min([y for (x, y, w, h) in bounding_boxes])
+        max_x = max([x + w for (x, y, w, h) in bounding_boxes])
+        max_y = max([y + h for (x, y, w, h) in bounding_boxes])
+        
+        # Scale the merged bounding box coordinates back to the original image size
+        min_x_scaled = int(min_x * scale_x)
+        min_y_scaled = int(min_y * scale_y)
+        max_x_scaled = int(max_x * scale_x)
+        max_y_scaled = int(max_y * scale_y)
+        
+        # Draw the merged bounding box on the original image
+        cv2.rectangle(img_with_boxes, (min_x_scaled, min_y_scaled), (max_x_scaled, max_y_scaled), (0, 255, 0), 2)
+    else:
+        img_with_boxes = img
+    
+    # Save the output image
+    output_path = 'output_image.jpg'
+    cv2.imwrite(output_path, img_with_boxes)
+
+    return img_with_boxes, prediction_label, prediction_status, recommendation, output_path
+
+
+interface = gr.Interface(
+    fn=predict,
+    inputs='image',
+    outputs=[
+        'image',
+        gr.components.Textbox(label='Disease Name'),
+        gr.components.Textbox(label='Disease status'),
+        gr.components.Textbox(label='Disease medicine'),
+        gr.components.File(label='Download Image')
+    ],
+    examples=[
+        ['download (1).jpeg'], ['download (2).jpeg'], ['download (3).jpeg'],
+        ['images (1).jpeg'], ['images (2).jpeg'], ['images (3).jpeg']
+    ]
+)
+interface.launch(debug=True)

fowl_pox_model.keras

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:43fd84c267a4c486c1819b245c544f0508e72576e1ec276faea9c43b8d82e43a
+size 218070017

requirements.txt

@@ -0,0 +1,3 @@
+tensorflow
+keras
+opencv-python