Merge pull request #1630 from Niraj1608/mask-detection

[Feature]: Streamlit Deployment and GUI for Face Mask Detection #1617

Detection Models/Face_Mask_Detection/Face_Mask_Detection.ipynb

@@ -1062,12 +1062,23 @@
    "id": "dbfeb3a0-d0a7-4446-8ae4-559206d46034",
    "metadata": {},
    "outputs": [],
+   "source": [
+    "#save the model\n",
+    "model.save('mask_detection_model.keras')\n"
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "id": "a62eb7fc",
+   "metadata": {},
+   "outputs": [],
    "source": []
   }
  ],
  "metadata": {
   "kernelspec": {
-   "display_name": "Python 3 (ipykernel)",
+   "display_name": "Python 3",
    "language": "python",
    "name": "python3"
   },
@@ -1081,7 +1092,7 @@
    "name": "python",
    "nbconvert_exporter": "python",
    "pygments_lexer": "ipython3",
-   "version": "3.11.5"
+   "version": "3.12.1"
   }
  },
  "nbformat": 4,

Detection Models/Face_Mask_Detection/predictive_system_gui.py

@@ -0,0 +1,83 @@
+import cv2
+import numpy as np
+import keras
+import tk as tk
+from PIL import ImageTk, Image
+import tkinter as tk
+from tkinter import filedialog
+import os
+
+# Load the face mask detector model
+model =keras.models.load_model("mask_detection.keras")
+
+# Initialize the GUI
+root = tk.Tk()
+root.title("Mask Detection")
+root.geometry("500x400")
+
+# Function to classify the image
+def classify_image(image):
+    # Preprocess the image
+    image = image.resize((128, 128))
+    image = np.array(image)
+    image = image / 255.0
+    input_image_reshaped = np.reshape(image, [1,128, 128, 3])
+    # Make predictions
+    predictions = model.predict(input_image_reshaped)
+    input_pred_label = np.argmax(predictions)
+
+    if input_pred_label == 1:
+        result = "With Mask"
+    else:
+        result = "Without Mask"
+
+    # Update the result label text
+    result_label.configure(text="Prediction: " + result)
+
+# Function to handle file upload
+def upload_image():
+    file_path = filedialog.askopenfilename(initialdir=os.getcwd(), title="Select Image File",
+                                               filetypes=(("JPEG files", "*.jpg"), ("PNG files", "*.png")))
+    if file_path:
+        # Display the uploaded image in the GUI
+        image = Image.open(file_path)
+        image.thumbnail((300, 300))
+        photo = ImageTk.PhotoImage(image)
+        photo_label.configure(image=photo)
+        photo_label.image = photo
+
+        classify_image(image)
+
+# Function to handle capturing photo from webcam
+def capture_photo():
+    video_capture = cv2.VideoCapture(0)
+    _, frame = video_capture.read()
+    video_capture.release()
+    frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
+    image = Image.fromarray(frame)
+    classify_image(image)
+
+    # Display the captured photo in the GUI
+    image.thumbnail((300, 300))
+    photo = ImageTk.PhotoImage(image)
+    photo_label.configure(image=photo)
+    photo_label.image = photo
+
+result_label = tk.Label(root, text="", font=("Helvetica", 16))
+result_label.pack(pady=10)
+
+# Create the GUI components
+upload_button = tk.Button(root, text="Upload Image", command=upload_image)
+upload_button.pack(pady=10)
+
+capture_button = tk.Button(root, text="Capture Photo", command=capture_photo)
+capture_button.pack(pady=10)
+
+result_label = tk.Label(root, text="")
+result_label.pack(pady=10)
+
+photo_label = tk.Label(root)
+photo_label.pack()
+
+# Run the GUI main loop
+root.mainloop()

Detection Models/Face_Mask_Detection/streamlit_app.py

@@ -0,0 +1,72 @@
+import cv2
+import numpy as np
+import keras
+import streamlit as st
+import PIL.Image
+
+model=keras.models.load_model('mask_detection.keras')
+
+
+
+
+
+def main():
+    st.title("Mask Detection App")
+    st.write("Please choose an option")
+
+    option = st.selectbox("Select an option", ("Upload Image", "Capture Photo"))
+
+    if option == "Upload Image":
+        uploaded_file = st.file_uploader("Choose an image", type=["jpg", "jpeg", "png"])
+
+        if uploaded_file is not None:
+            # Read and preprocess the image
+            image = PIL.Image.open(uploaded_file)
+            image = image.resize((128,128))
+            image = np.array(image)  # Convert to NumPy array
+            image = image / 255.0  # Normalize the image
+            input_image_reshaped = np.reshape(image, [1, 128, 128, 3])
+
+            # Make predictions
+            predictions = model.predict(input_image_reshaped)
+            input_pred_label = np.argmax(predictions)
+
+            # Display the result
+            if input_pred_label == 1:
+                st.markdown("<h3 style='text-align: center; '>Prediction: With Mask</h3>", unsafe_allow_html=True)
+            else:
+                st.markdown("<h3 style='text-align: center; '>Prediction: Without Mask</h3>", unsafe_allow_html=True)
+
+            # Display the uploaded image
+            st.write("")
+            st.write("**Uploaded Image**")
+            st.image(image, use_column_width=True)
+
+    elif option == "Capture Photo":
+        video_capture = cv2.VideoCapture(0)
+        _, frame = video_capture.read()
+        video_capture.release()
+
+        # Convert captured frame to RGB
+        frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
+
+        # Resize and preprocess the image
+        image = cv2.resize(frame, (64, 64))
+        image = image / 255.0
+        input_image_reshaped = np.reshape(image, [1, 64, 64, 3])
+
+        # Make predictions
+        predictions = model.predict(input_image_reshaped)
+        input_pred_label = np.argmax(predictions)
+
+        # Display the result
+        if input_pred_label == 1:
+            st.markdown("<h3 style='text-align: center; '>Prediction: With Mask</h3>", unsafe_allow_html=True)
+        else:
+            st.markdown("<h3 style='text-align: center; '>Prediction: Without Mask</h3>", unsafe_allow_html=True)
+
+        # Display the captured photo
+        st.image(frame, channels="RGB", use_column_width=True)
+
+if __name__ == '__main__':
+    main()