Merge pull request larymak#137 from erastusnzula/face_recognition

Face recognition

Author: larymak

.gitignore

@@ -133,3 +133,4 @@ cython_debug/
 !.vscode/launch.json 
 !.vscode/extensions.json 
 .history
+.idea

Face_recognition/.gitignore

@@ -0,0 +1 @@
+.idea

Face_recognition/images/erastus.png

@@ -0,0 +1,115 @@
+import os
+import winsound
+
+import cv2
+import face_recognition
+import numpy
+
+
+class FaceRecognition:
+
+    def __init__(self):
+        self.images = []
+        self.base_names = []
+        self.encoded_faces = []
+        self.path = 'images'
+        print('please wait my dummy program is busy encoding...')
+        self.camera = cv2.VideoCapture(0, cv2.CAP_DSHOW)
+        self.known_encoded_faces = []
+        self.frame_one = None
+        self.video_capture = None
+
+    def save_video_stream(self):
+        os.makedirs('Security Camera Video', exist_ok=True)
+        name = 'Video1.avi'
+        if name in os.listdir('Security Camera Video'):
+            name = f'{os.path.splitext(name)[0][:-1]}' \
+                   f'{len(os.listdir("Security Camera Video")) + 1}' \
+                   f'{os.path.splitext(name)[1]}'
+        fourcc = cv2.VideoWriter_fourcc(*'XVID')
+        self.video_capture = cv2.VideoWriter(os.path.join('Security Camera Video', name),
+                                             fourcc, 20.0, (640, 480))
+        print(f'{name} saving ...')
+
+    def known_image_folder(self):
+        for image in os.listdir(path=self.path):
+            if not (image.endswith('.jpeg') or image.endswith('.JPG')
+                    or image.endswith('.jpg') or image.endswith(
+                    '.png') or image.endswith('.PNG')):
+                continue
+            img = cv2.imread(f"{self.path}/{image}")
+            self.images += [img]
+            self.base_names += [os.path.splitext(image)[0]]
+
+    def encode_face(self, images):
+        for image in images:
+            image_rgb = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
+            face_encoding = face_recognition.face_encodings(image_rgb)[0]
+            self.encoded_faces += [face_encoding]
+        return self.encoded_faces
+
+    def run(self):
+        self.known_image_folder()
+        self.known_encoded_faces = self.encode_face(self.images)
+        print("Encoding Complete, camera opening...")
+        self.save_video_stream()
+        self.open_camera_tasks()
+
+    def open_camera_tasks(self):
+        while self.camera.isOpened():
+            _, self.frame_one = self.camera.read()
+            self.detect_faces()
+            self.video_capture.write(self.frame_one)
+            cv2.imshow("Press 'c' to capture image, q to quit.", self.frame_one)
+            key = cv2.cv2.waitKey(1)
+            if key == ord('q') or key == 27:
+                self.exit_protocol()
+            elif key == ord('c') or key == 32:
+                self.save_image_capture()
+
+    def save_image_capture(self):
+        os.makedirs('Security Camera Images', exist_ok=True)
+        name = 'Image1.png'
+        if name in os.listdir('Security Camera Images'):
+            name = f"{os.path.splitext(name)[0][:-1]}\
+            {len(os.listdir('Security Camera Images')) + 1}" \
+                   f"{os.path.splitext(name)[1]}"
+        cv2.imwrite(os.path.join('Security Camera Images', name), self.frame_one)
+        print(f"{name} captured successfully.")
+
+    def detect_faces(self):
+        frame_one_rgb = cv2.cvtColor(self.frame_one, cv2.COLOR_BGR2RGB)
+        face_location = face_recognition.face_locations(frame_one_rgb)
+        face_encoding = face_recognition.face_encodings(frame_one_rgb, face_location)
+        for face_encoded, face_located in zip(face_encoding, face_location):
+            self.face_comparison(face_encoded, face_located)
+
+    def face_comparison(self, face_encoded, face_located):
+        compare_face = face_recognition.compare_faces(self.known_encoded_faces, face_encoded)
+        face_distance = face_recognition.face_distance(self.known_encoded_faces, face_encoded)
+        distance_index = numpy.argmin(face_distance)
+        if compare_face[distance_index]:
+            name = self.base_names[int(distance_index)].title()
+            x, y, w, h = face_located
+            cv2.rectangle(self.frame_one, (h, x), (y, w), (0, 255, 0), 2)
+            cv2.rectangle(self.frame_one, (h, x-40), (y, w), (0, 255, 0), 2)
+            cv2.putText(self.frame_one, name, (
+                h + 30, x - 10), cv2.FONT_HERSHEY_COMPLEX_SMALL, 1, (0, 255, 255), 2)
+        else:
+            x, y, w, h = face_located
+            cv2.rectangle(self.frame_one, (h, x), (y, w), (0, 255, 0), 2)
+            cv2.rectangle(self.frame_one, (h, x - 40), (y, w), (0, 255, 0), 2)
+            cv2.putText(self.frame_one, "Unknown", (h, x - 10),
+                        cv2.FONT_HERSHEY_COMPLEX_SMALL, 1, (0, 255, 255),
+                        2)
+            winsound.PlaySound('alert.wav', winsound.SND_ASYNC)
+            print("Intruder in the house.")
+
+    def exit_protocol(self):
+        self.camera.release()
+        self.video_capture.release()
+        cv2.destroyAllWindows()
+
+
+if __name__ == "__main__":
+    FaceRecognition().run()

Face_recognition/main.py

@@ -0,0 +1,26 @@
+Face recognition
+================
+
+Description
+===========
+```
+* Detect faces in a live camera feed. 
+* Compare them to directory images.
+* Attach a name tag.
+```
+Packages
+========
+
+```
+* import os.
+* import winsound.
+* import cv2.
+* import face_recognition.
+* import numpy.
+```
+How it works
+============
+```
+* Clone the project.
+* Run main.py.
+```

Face_recognition/readme.md

@@ -95,4 +95,5 @@ The contribution guidelines are as per the guide [HERE](https://github.com/larym
 | 51    | [PDF Downloader](https://github.com/Sdccoding/Python-project-Scripts/tree/main/PDF_Downloader)                                       | [Souhardya Das Chowdhury](https://github.com/Sdccoding)
 | 52    | [ConsoleSnake](https://github.com/larymak/Python-project-Scripts/tree/main/ConsoleSnake)                                       | [tomimara52](https://github.com/tomimara52)
 | 52    | [ConsoleMinesweeper](https://github.com/larymak/Python-project-Scripts/tree/main/ConsoleMinesweeper)                                       | [tomimara52](https://github.com/tomimara52)
-| 53    | [Drink Water Reminder](https://github.com/larymak/Python-project-Scripts/tree/main/drink-water-reminder)                                       | [Tristán Hdez](https://github.com/tristanHdez18)
+
+| 53    | [Face_recognition](https://github.com/erastusnzula/Python-project-Scripts/tree/face_recognition/Face_recognition) |[erastusnzula](https://github.com/erastusnzula)