Update app.py

app.py

@@ -1,108 +1,36 @@
-import gradio as gr
-
-import cv2
-from mtcnn.mtcnn import MTCNN
-import tensorflow as tf
-import tensorflow_addons
-import numpy as np
-
-import os
-#import zipfile
-
-
-#local_zip = "FINAL-EFFICIENTNETV2-B0.zip"
-#zip_ref = zipfile.ZipFile(local_zip, 'r')
-#zip_ref.extractall('FINAL-EFFICIENTNETV2-B0')
-#zip_ref.close()
-
-import h5py
-
-# افتح ملف النموذج H5
-h5_file = h5py.File('model_CNN.h5', 'r')
-
-# قراءة المحتوى
-# يمكنك استكشاف المحتوى باستخدام الأوامر التالية
-print(h5_file.keys())  # قائمة المفاتيح في الملف
-print(h5_file['model_weights'].keys())  # قائمة المفاتيح في مجموعة الأوزان
-
-# إنشاء ملف HDF5 جديد
-hdf5_file = h5py.File('model.hdf5', 'w')
-
-# نسخ المحتوى من ملف النموذج H5 إلى الملف الجديد
-h5_file.copy('model_weights', hdf5_file)
-
-# إغلاق الملفات
-h5_file.close()
-hdf5_file.close()
 
-model = tf.keras.models.load_model('model.hdf5')
 
-detector = MTCNN()
-
-def deepfakespredict(input_img ):
-
-    labels = ['real', 'fake']
-    pred = [0, 0]
-    text =""
-    text2 =""
-    
-    face = detector.detect_faces(input_img)
-
-    if len(face) > 0:
-        x, y, width, height = face[0]['box']
-        x2, y2 = x + width, y + height
-
-        cv2.rectangle(input_img, (x, y), (x2, y2), (0, 255, 0), 2)
-
-        face_image = input_img[y:y2, x:x2]
-        face_image2 = cv2.cvtColor(face_image, cv2.COLOR_BGR2RGB)
-        face_image3 = cv2.resize(face_image2, (224, 224))
-        face_image4 = face_image3/255
-
-        pred = model.predict(np.expand_dims(face_image4, axis=0))[0]
+import tensorflow_hub as hub
+import gradio as gr
+import tensorflow as tf
 
-        if pred[1] >= 0.6:
-            text = "The image is FAKE."
-        elif pred[0] >= 0.6:
-            text = "The image is REAL."
-        else:
-            text = "The image may be REAL or FAKE."
-            
+# Load the model
+model = tf.keras.models.load_model('model_CNN')
+
+# Preprocess the image
+def preprocess_image(image):
+    resized_image = tf.image.resize(image, [224, 224])
+    normalized_image = resized_image / 255.0
+    return normalized_image
+
+# Function to detect fake images using the model
+def detect_fake_image(image):
+    processed_image = preprocess_image(image)
+    prediction = model.predict(tf.expand_dims(processed_image, 0))
+    percentage = prediction[0][0] * 100
+    if percentage > 50:
+        result = f"Real with {percentage}% confidence"
     else:
-        text = "Face is not detected in the image."
-    
-    text2 = "REAL: " + str(np.round(pred[0]*100, 2)) + "%, FAKE: " + str(np.round(pred[1]*100, 2)) + "%"
-
-    return input_img, text, text2, {labels[i]: float(pred[i]) for i in range(2)}
-
-
-title="Deepfakes Image Detector"
-description=" project Deepfake detection images real and fake call me +967776215118  "
-            
-examples = [
-                ['Fake-1.png'],
-                ['Fake-2.png'],
-                ['Fake-3.png'],
-                ['Fake-4.png'],
-                ['Fake-5.png'],
-  
-                ['Real-1.png'],
-                ['Real-2.png'],
-                ['Real-3.png'],
-                ['Real-4.png'],
-                ['Real-5.png']
-                
-           ]
-           
-             
-gr.Interface(deepfakespredict,
-                     inputs = ["image"],
-                     outputs=[gr.outputs.Image(type="pil", label="Detected face"), 
-                         "text", 
-                         "text", 
-                         gr.outputs.Label(num_top_classes=None, type="auto", label="Confidence")],                     
-                     title=title,
-                     description=description,
-                     examples = examples, 
-                     examples_per_page = 5
-                     ).launch()
+        result = f"Fake with {100 - percentage}% confidence"
+    return result
+
+# Gradio Interface
+iface = gr.Interface(
+    fn=detect_fake_image,
+    inputs="image",
+    outputs="text",
+    title="Fake Image Detector"
+)
+
+# Launch the interface
+iface.launch()