📌 Face Recognition Attendance System

📍Overview

This project implements a face recognition-based attendance system using Python libraries. The system captures video from a webcam, recognizes faces, and updates attendance information in csv file.

Setting up the Development Environment
Collecting Training Data
Preprocessing and Training the Face Recognition Model
Implementing the Attendance System
Challenges Faced by Face Recognition System

📍Setting up the Development Environment

Visit the official Python website to download the Python version 3.6.8 on your local system.
Download C++ Desktop development tools here.
Create a empty folder and run the following command in the terminal git clone https://github.com/harshpreet14/IDC_409_ImageProcessing.git
Next, run pip install -r requirements.txt

To build our face recognition attendance system, we will need the following Python libraries:

📍OpenCV: For computer vision tasks such as face detection and image processing.

📍Dlib: A powerful library that provides facial landmark detection and face alignment capabilities.

📍face_recognition: A Python library built on top of dlib, designed specifically for face recognition tasks.

📍Pandas: For managing and analyzing the attendance data.

NOTE- Please make sure to use Python version 3.6.8 and exact versions of libraries according to requirements.txt, else the code might not work due to version incompatibility.

📍Collecting Training Data

Before training the face recognition model, we need a dataset of labeled images representing different individuals. Start by gathering images of each person to be recognized. Store the images in Images folder in root directory.

📍Preprocessing and Training the Face Recognition Model

To train the face recognition model, we will use a popular pre-trained model called "dlib_face_recognition_resnet_model_v1." This model provides a 128-dimensional face embedding for each face detected in an image.

Load images After importing libraries you need to load an image. By default, face_recognition library loads images in the form of BGR, in order to print the image you should convert it into RGB using OpenCV.

# Importing the student images
folderPath = 'Images'
pathList = os.listdir(folderPath)
imgList = []
studentIds =[]
for path in pathList:
    imgList.append(cv2.imread(os.path.join(folderPath, path)))
    studentIds.append(os.path.splitext(path)[0])

Find Encodings for Images

def findEncodings(imagesList):
    encodeList=[]
    for img in imagesList:
        img = cv2.cvtColor(img,cv2.COLOR_BGR2RGB)
        encode = face_recognition.face_encodings(img)[0]
        encodeList.append(encode)

    return encodeList

3.Save Encodings as pickle file

file = open("EncodeFile.p", "wb")
pickle.dump(encodeListKnownWithIds, file)
file.close()
print("File Saved")

📍Implementing the Attendance System

Now that we have trained our face recognition model, we can proceed to implement the attendance system. The system will follow these steps:

Capture a live video stream using OpenCV

cap = cv2.VideoCapture(0)
cap.set(3, 640)
cap.set(4, 480)

Detect faces in each frame using the face_recognition library

    success, img = cap.read()

    imgS = cv2.resize(img, (0, 0), None, 0.25, 0.25)
    imgS = cv2.cvtColor(imgS, cv2.COLOR_BGR2RGB)

    faceCurFrame = face_recognition.face_locations(imgS)

Compute face embeddings for the detected faces

encodeCurFrame = face_recognition.face_encodings(imgS, faceCurFrame)

Compare the computed embeddings with the embeddings of known individuals

if faceCurFrame:
        for encodeFace, faceLoc in zip(encodeCurFrame, faceCurFrame):
            matches = face_recognition.compare_faces(encodeListKnown, encodeFace)
            faceDis = face_recognition.face_distance(encodeListKnown, encodeFace)
            # print("matches", matches)
            # print("faceDis", faceDis)

            matchIndex = np.argmin(faceDis)
            # print("Match Index", matchIndex)

            if matches[matchIndex]:
                # print("Known Face Detected")
                # print(studentIds[matchIndex])
                y1, x2, y2, x1 = faceLoc
                y1, x2, y2, x1 = y1 * 4, x2 * 4, y2 * 4, x1 * 4
                bbox = 55 + x1, 162 + y1, x2 - x1, y2 - y1
                imgBackground = cvzone.cornerRect(imgBackground, bbox, rt=0)
                student_id = studentIds[matchIndex]

If a match is found, mark the attendance for that person

 # Save updated data back to CSV
                        with open("students.csv", 'a') as csvfile:
                            fieldnames = ['id', 'last_attendance_time', 'total_attendance']
                            writer = csv.DictWriter(csvfile, fieldnames=fieldnames)
                            writer.writerow(studentInfo)

📍Challenges Faced by Face Recognition System

🎯Illumination: It changes the face’s appearance drastically. It is observed that even slight changes in lighting conditions cause a significant impact on its results.
🎯Pose: Facial Recognition systems are highly sensitive to the pose, Which may result in faulty recognition or no recognition if the database is only trained on frontal face view.
🎯Facial Expressions: Different expressions of the same individual are another significant factor that needs to be taken into account. Modern Recognizers can easily deal with it, though.
🎯Low Resolution: The recognizer must be trained on a good-resolution picture. Otherwise, the model will fail to extract features.
🎯Aging: With increasing age, the human face features shape, lines, and texture changes which are yet another challenge.

harshpreet14 / IDC_409_ImageProcessing