Add detector.py file

Author: MaheshRKumawat

README.md

@@ -6,7 +6,7 @@ You can download the weights of all models and h5 file of GAN_fake_vs_real model
 After downloading the repository, you can run the requirements.txt file to install the required packages.
 <br>
 
-You can run the predict.ipynb file from each directory to test the model.
+You can run the predict.ipynb file from each directory to test the model. 
 
 ### Note: 
 If you get this error `ImportError: cannot import name 'Sequence' from 'keras.utils' (/usr/local/lib/python3.7/dist-packages/keras/utils/__init__.py)`
@@ -15,9 +15,17 @@ then open the file in location `/usr/local/lib/python3.7/dist-packages/keras_vid
 and then replace line `from keras.utils import Sequence` with `from tensorflow.keras.utils import Sequence` and save it.
 <br>
 
+### Detector.py file:
+It consists of function that can be used to detect deepfake or ganfake etc.
+<br>
+you can call the function `predict` which accepts two parameters:
+1. path of the video or image
+2. type which you want to check (deepfake video, deepfake image, ganfake)
+
 ## Directory Structure
 ```
 |   .gitignore
+|   detector.py
 |   README.md
 |   requirements.txt
 |

detector.py

@@ -0,0 +1,130 @@
+import os
+import cv2
+import numpy as np
+import tensorflow as tf
+from tensorflow import keras
+from tensorflow.keras.preprocessing import image
+
+
+def predict_deepfake_video(video_path, model_path, model_weights_path, faceCascade_path):
+    """
+    Predict if a video is a deepfake or not.
+    """
+
+    # Load model and weights
+    model = keras.models.load_model(model_path, compile=False)
+    model.load_weights(model_weights_path)
+
+    cap = cv2.VideoCapture(video_path)
+    img_array = []
+    success = 1
+    i = 1
+    faceCascade = cv2.CascadeClassifier(faceCascade_path)
+
+    while success and i <= 90:
+        success, img = cap.read()
+        try:
+            imgGray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
+            faces = faceCascade.detectMultiScale(imgGray, 1.1, 4)
+            if np.shape(faces) == (1, 4):
+                x, y, w, h = faces[0]
+                imgCropped = img[y: y + h, x:x + w]
+                imgCropped = cv2.resize(imgCropped, (112, 112))
+                img_array.append(imgCropped)
+                i += 1
+        except:
+            return "Provide a video with more than 90 frames"
+
+    img_array = np.array(img_array)
+    img_array = img_array.reshape(1, 90, 112, 112, 3)
+
+    res = model.predict(img_array).round()
+    if res[0][0] == 0:
+        return "Real"
+    else:
+        return "Deepfake"
+
+
+def predict_deepfake_image(image_path, model_path, model_weights_path):
+    """
+    Predict if an image is a deepfake or not.
+    """
+
+    model = keras.models.load_model(model_path, compile=False)
+    model.load_weights(model_weights_path)
+    img = image.load_img(image_path, target_size=(256, 256))
+    x = image.img_to_array(img)
+    x = np.expand_dims(x, axis=0)
+    x = tf.image.resize(x, (256, 256))
+    x /= 255.0
+    res = model.predict(x)
+    if(res[0][0] < 0):
+        return "Deepfake"
+    else:
+        return "Real"
+
+
+def predict_gan_fake(image_path, model_path, model_weights_path):
+    """
+    Predict if an image is a GAN Fake or Real.
+    """
+
+    model = keras.models.load_model(model_path, compile=False)
+    model.load_weights(model_weights_path)
+    img = image.load_img(image_path, target_size=(256, 256))
+    x = image.img_to_array(img)
+    x = np.expand_dims(x, axis=0)
+    x = tf.image.resize(x, (256, 256))
+    x /= 255
+
+    res = model.predict(x)
+
+    if(res[0][0] < 0):
+        return "GAN Fake"
+    else:
+        return "Real"
+
+
+def predict(input_path, type):
+    """
+    It takes input and type
+    Based on the type it does the prediction on input
+    "1" for deepfake video
+    "2" for deepfake image
+    "3" for GAN Fake
+    """
+
+    physical_devices = tf.config.list_physical_devices('GPU')
+    tf.config.experimental.set_memory_growth(physical_devices[0], True)
+
+    # check if input path exists and is a file
+    if not os.path.exists(input_path) or not os.path.isfile(input_path):
+        return "Provide a valid path"
+
+    if type == "1":
+        model_path = "deepfake_video/video_model.h5"
+        model_weights_path = "deepfake_video/Weights/model_weights"
+        faceCascade_path = "deepfake_video/Resources/haarcascade_frontalface_default.xml"
+        try:
+            return predict_deepfake_video(input_path, model_path, model_weights_path, faceCascade_path)
+        except:
+            return "No video found"
+
+    elif type == "2":
+        model_path = "deepfake_image\CNN_SVM_Model\cnn_svm_model.h5"
+        model_weights_path = "deepfake_image\Weights_CNN_SVM\model_weights"
+        try:
+            return predict_deepfake_image(input_path, model_path, model_weights_path)
+        except:
+            return "No image found"
+
+    elif type == "3":
+        model_path = "GAN_Fake_vs_Real/Inception_Resnet_SVM_Model\Inception_Resnet_svm_model.h5"
+        model_weights_path = "GAN_Fake_vs_Real/Weights_Inception_Resnet_SVM/model_weights"
+        try:
+            return predict_gan_fake(input_path, model_path, model_weights_path)
+        except:
+            return "No image found"
+
+    else:
+        return "Invalid type"