Recognization Feature Works. Fixed Bugs as well ;)

.gitignore

@@ -71,5 +71,8 @@ fabric.properties
 .idea/caches/build_file_checksums.ser
 
 # our project
+Attendance/
+training_images/
+unknown_images/
+files/trainer.yml
 trainer.yml
-training_images/

files/student_details.csv

@@ -1,4 +1,4 @@
-Id,Name
+ID,Name
 ,
 ,
 ,
@@ -17,5 +17,4 @@ Id,Name
 ,
 ,
 ,
-0, Test
-1,Aslam
+0,Test

main.py

@@ -6,6 +6,16 @@
 import pyfiglet
 
 
+# input live stream from a recorder
+INPUT_VIDEO = "http://192.168.1.100:8080/video"
+
+# input from saved video
+# INPUT_VIDEO = "video.avi"
+
+# input from a device attached to computer
+# INPUT_VIDEO = 0  # or -1 if 0 doesn't work
+
+
 # creating the title bar function
 def title_bar():
     # os.system('cls')  # for windows
@@ -31,16 +41,16 @@ def main_menu():
             choice = int(input("Enter Choice: "))
 
             if choice == 1:
-                check_camera()
+                check_camera(INPUT_VIDEO)
                 break
             elif choice == 2:
-                capture_face()
+                capture_face(INPUT_VIDEO)
                 break
             elif choice == 3:
                 train_face()
                 break
             elif choice == 4:
-                recognize_face()
+                recognize_face(INPUT_VIDEO)
                 break
             elif choice == 5:
                 print("Thank You =)")
@@ -57,15 +67,15 @@ def main_menu():
 
 # ---------------------------------------------------------
 # calling the camera test function from check camera.py file
-def check_camera():
-    capture_video.start()
+def check_camera(input_video):
+    capture_video.start(input_video)
     main_menu()
 
 
 # --------------------------------------------------------------
 # calling the take image function form capture image.py file
-def capture_face():
-    capture_images.capture()
+def capture_face(input_video):
+    capture_images.capture(input_video)
     main_menu()
 
 
@@ -78,8 +88,8 @@ def train_face():
 
 # --------------------------------------------------------------------
 # calling the recognize_attendance from recognize.py file
-def recognize_face():
-    recognize.mark_attendance()
+def recognize_face(input_video):
+    recognize.mark_attendance(input_video)
     main_menu()
 
 

modules/capture_images.py

@@ -3,15 +3,6 @@
 import cv2
 import unicodedata  # to check if entered in different unicode format
 
-# input live stream from a recorder
-# URL = "http://192.168.1.103:8080/video"
-
-# input from saved video
-# URL = "video.avi"
-
-# input from a device attached to computer
-URL = 0  # or -1
-
 
 # check if student_id is a number
 def is_number(_id):
@@ -31,13 +22,14 @@ def is_number(_id):
 
 
 # Capture Image function definition
-def capture():
+def capture(input_video):
     student_id = input("Enter Your ID (numbers only): ")
     name = input("Enter Your Name (alphabets only): ")
 
     # if "student_id is a number" and "name consists of alphabetic chars only" then
     if is_number(student_id) and name.isalpha():
-        cap = cv2.VideoCapture(URL)
+        # store input video stream in cap variable
+        cap = cv2.VideoCapture(input_video)
 
         # using haar cascade
         haar_cascade_path = "files" + os.sep + "haarcascade_frontalface_default.xml"
@@ -48,35 +40,38 @@ def capture():
         while True:
             # capture frame-by-frame
             ret, img = cap.read()
-
-            if ret is True:
-                # operations on the frame come here
+            if ret is True:  # video is detected
+                # convert frame to grayscale
                 gray_frame = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
 
+                # detect faces using haar cascade detector
                 faces = detector.detectMultiScale(gray_frame, 1.3, 5)
                 for(x, y, w, h) in faces:
-                    cv2.rectangle(gray_frame, (x, y), (x+w, y+h), (255, 0, 0), 2)  # ##gray
+                    cv2.rectangle(img, (x, y), (x+w, y+h), (255, 0, 0), 2)  # ##gray
 
                     # incrementing number
                     increment_num += 1
 
                     # saving the captured face in the data-set folder training_images
                     cv2.imwrite("training_images" + os.sep + name + "." + student_id + '.' +
-                                str(increment_num) + ".jpg", gray_frame[y:y+h, x:x+w])  # ##gray[y:y+h, x:x+w]
+                                str(increment_num) + ".jpg", img[y:y+h, x:x+w])  # ##gray[y:y+h, x:x+w]
 
                     # display the resulting frame
-                    cv2.imshow('frame', gray_frame)  # ##gray
+                    cv2.imshow('Capturing Face - Attendance using Face Recognition', img)  # ##gray
 
                 # wait for 100 milliseconds
                 if cv2.waitKey(100) & 0xFF == ord('q'):
                     break
                 # break if the sample number is more than 100
                 elif increment_num > 60:
                     break
-            else:
+            else:  # video not detected
                 break
+
+        # when everything is done
         cap.release()
         cv2.destroyAllWindows()
+
         # res = "Images Saved for ID : " + student_id + " Name : " + name
         row = [student_id, name]
         with open("files" + os.sep + "student_details.csv", 'a+') as csv_file:

modules/capture_video.py

@@ -1,30 +1,25 @@
 import cv2
 
-# input live stream from a recorder
-# URL = "http://192.168.1.103:8080/video"
 
-# input from saved video
-# URL = "video.avi"
+def start(input_video):
+    # store input video stream capture in cap variable
+    cap = cv2.VideoCapture(input_video)
 
-# input from a device attached to computer
-URL = 0  # or -1
-
-
-def start():
-    cap = cv2.VideoCapture(URL)
     while cap.isOpened():
         # capture frame-by-frame
         ret, frame = cap.read()
         if ret is True:  # video is detected
+            # convert frame to grayscale
             # gray_frame = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
 
             # display the resulting frame
-            cv2.imshow('frame', frame)
+            cv2.imshow('Checking Video - Attendance using Face Recognition', frame)
 
             if cv2.waitKey(1) & 0xFF == ord('q'):
                 break
         else:  # video not detected
             break
+
     # when everything is done
     cap.release()
     cv2.destroyAllWindows()

modules/recognize.py

@@ -5,53 +5,72 @@
 import pandas as pd
 
 
-def mark_attendance():
+def mark_attendance(input_video):
+    # reading trained dataset
     recognizer = cv2.face.LBPHFaceRecognizer_create()  # cv2.createLBPHFaceRecognizer()
-    recognizer.read("TrainingImageLabel"+os.sep+"trainer.yml")
-    harcascadePath = "haarcascade_frontalface_default.xml"
-    faceCascade = cv2.CascadeClassifier(harcascadePath)
-    df = pd.read_csv("StudentDetails"+os.sep+"student_details.csv")
-    cam = cv2.VideoCapture(0)
+    recognizer.read("files" + os.sep + "trainer.yml")
+
+    # using haar cascade
+    haar_cascade_path = "files" + os.sep + "haarcascade_frontalface_default.xml"
+    face_cascade = cv2.CascadeClassifier(haar_cascade_path)
+
+    # preparing pandas dataframe
+    df = pd.read_csv("files" + os.sep + "student_details.csv")
+    col_names = ['ID', 'Name', 'Date', 'Time']
+    attendance_df = pd.DataFrame(columns=col_names)
+
+    # store input video stream capture in cap variable
+    cam = cv2.VideoCapture(input_video)
     font = cv2.FONT_HERSHEY_SIMPLEX
-    col_names = ['Id', 'Name', 'Date', 'Time']
-    attendance = pd.DataFrame(columns=col_names)
 
     while True:
-        ret, im = cam.read()
-        gray = cv2.cvtColor(im, cv2.COLOR_BGR2GRAY)
-        faces = faceCascade.detectMultiScale(gray, 1.2, 5)
-        for(x, y, w, h) in faces:
-            cv2.rectangle(im, (x, y), (x+w, y+h), (225, 0, 0), 2)
-            Id, conf = recognizer.predict(gray[y:y+h, x:x+w])
-
-            if(conf < 50):
-                ts = time.time()
-                date = datetime.datetime.fromtimestamp(ts).strftime('%Y-%m-%d')
-                timeStamp = datetime.datetime.fromtimestamp(
-                    ts).strftime('%H:%M:%S')
-                aa = df.loc[df['Id'] == Id]['Name'].values
-                tt = str(Id)+"-"+aa
-                attendance.loc[len(attendance)] = [Id, aa, date, timeStamp]
-
-            else:
-                Id = 'Unknown'
-                tt = str(Id)
-            if(conf > 75):
-                noOfFile = len(os.listdir("ImagesUnknown"))+1
-                cv2.imwrite("ImagesUnknown"+os.sep+"Image"+str(noOfFile) +
-                            ".jpg", im[y:y+h, x:x+w])
-            cv2.putText(im, str(tt), (x, y+h), font, 1, (255, 255, 255), 2)
-        attendance = attendance.drop_duplicates(subset=['Id'], keep='first')
-        cv2.imshow('im', im)
-        if (cv2.waitKey(1) == ord('q')):
+        # capture frame-by-frame
+        ret, img = cam.read()
+        if ret is True:  # video is detected
+            # convert frame to grayscale
+            gray_frame = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
+
+            # detect faces using haar cascade detector
+            faces = face_cascade.detectMultiScale(gray_frame, 1.2, 5)
+            for(x, y, w, h) in faces:
+                cv2.rectangle(img, (x, y), (x+w, y+h), (225, 0, 0), 2)
+                _id, conf = recognizer.predict(gray_frame[y:y+h, x:x+w])
+
+                if conf < 50:
+                    current_time = time.time()
+                    date = datetime.datetime.fromtimestamp(current_time).strftime('%Y-%m-%d')
+                    timestamp = datetime.datetime.fromtimestamp(current_time).strftime('%H:%M:%S')
+                    student_name = df.loc[df['ID'] == _id]['Name'].values[0]
+                    display_text = student_name  # str(_id) + "-" +
+                    attendance_df.loc[len(attendance_df)] = [_id, student_name, date, timestamp]
+
+                else:
+                    display_text = 'Unknown'
+
+                if conf > 75:
+                    file_number = len(os.listdir("unknown_images")) + 1
+                    cv2.imwrite("unknown_images" + os.sep + "Image" + str(file_number) +
+                                ".jpg", img[y:y+h, x:x+w])
+                cv2.putText(img, display_text, (x, y+h), font, 1, (255, 255, 255), 2)
+            attendance_df = attendance_df.drop_duplicates(subset=['ID'], keep='first')
+            cv2.imshow('Recognizing Faces - Attendance using Face Recognition', img)
+            if cv2.waitKey(1) == ord('q'):
+                break
+        else:  # video not detected
             break
-    ts = time.time()
-    date = datetime.datetime.fromtimestamp(ts).strftime('%Y-%m-%d')
-    timeStamp = datetime.datetime.fromtimestamp(ts).strftime('%H:%M:%S')
-    Hour, Minute, Second = timeStamp.split(":")
-    fileName = "Attendance"+os.sep+"Attendance_"+date+"_"+Hour+"-"+Minute+"-"+Second+".csv"
-    attendance.to_csv(fileName, index=False)
+
+    # get current time and date
+    current_time = time.time()
+    date = datetime.datetime.fromtimestamp(current_time).strftime('%Y-%m-%d')
+    timestamp = datetime.datetime.fromtimestamp(current_time).strftime('%H:%M:%S')
+    hour, minute, second = timestamp.split(":")
+
+    # create a csv(comma separated value) file and append current date and time to its name
+    file_name = "Attendance" + os.sep + "Attendance_" + date + "_" + hour + "-" + minute + "-" + second + ".csv"
+    attendance_df.to_csv(file_name, index=False)
+
+    # when everything is done
     cam.release()
     cv2.destroyAllWindows()
 
-    print("Attendance Successfull")
+    print("Attendance Successful!")

modules/train_images.py

@@ -32,9 +32,9 @@ def get_images_and_labels(path):
 # ----------- train images function ---------------
 def train():
     recognizer = cv2.face_LBPHFaceRecognizer.create()
-    # haar_cascade_path = "files"+os.sep+"haarcascade_frontalface_default.xml"
+    # haar_cascade_path = "files" + os.sep + "haarcascade_frontalface_default.xml"
     # detector = cv2.CascadeClassifier(haar_cascade_path)
     faces, _id = get_images_and_labels("training_images")
     recognizer.train(faces, np.array(_id))
-    recognizer.save("files"+os.sep+"trainer.yml")
+    recognizer.save("files" + os.sep + "trainer.yml")
     print("Images Trained Successfully")

requirements.txt

@@ -1,3 +1,4 @@
+pyfiglet
 appdirs==1.4.3
 attrs==19.1.0
 black==19.3b0