TECHREPORT.txt

*********************************************************TECHNICAL REPORT**********************************************************************


Required Libraries and Dependencies:
The program utilizes the following libraries and dependencies:
pyttsx3: Text-to-speech library for audio feedback.
face_recognition: Face recognition library for face detection and recognition.
cv2: OpenCV library for image and video processing.
numpy: Numerical computing library for array operations.
io: Input/output library for handling images and bytes.
speech_recognition: Speech recognition library for voice input processing.
Face Detection and Recognition:
The facedetect function captures video frames from the webcam, displays them, and detects faces using the face_recognition library. 
It compares the detected faces with known face encodings and names to identify individuals. If a match is found, the function returns 
the name of the recognized person.

Speech Recognition:
The textfromspeech function utilizes the speech_recognition library to capture speech input from the microphone. It adjusts for ambient noise, 
listens for audio input, and then uses the Google Speech Recognition API to convert the audio into text. The recognized text is returned as the output.

User Classes and Data Structures:
The program uses three classes: Student, Employee, and Visitor, each representing a different type of user. The program initializes lists 
of 1000 instances for each class to store the user objects. For each user, two known face images are associated with their name and stored 
in the known_image_paths list. The names are stored in the known_names list.

MySQL Implementation:
SQL implementation is also done to store the dataset of Students(only), so after adding the images in faces/ folder, the students can be added in the 
MySQL database through the commands given in README.md, so even if the code face some errors or bugs or stop functioning, the database wont be lost and code can be run again and
all the previous details such as inside outside status or visit place(if outside) will continue to be the same.

User Identification and Access Control:
The object_type function determines the user type (student, employee, or visitor) based on the input name. It searches through the respective 
user lists and sets the global variables user_type and user_index to identify the user.

Access Control Logic:
The main function serves as the entry point of the program. It calls the facedetect function to capture the face of the individual 
seeking access. The object_type function is then called to identify the user. Based on the user type and access status, the program 
provides appropriate audio feedback and performs access control actions:

For students, the program asks for the desired destination, updates the visit place and inside status, and provides audio feedback accordingly.
For employees, the program greets them and updates the inside status based on their previous status.
For visitors, the program provides instructions and updates the inside status accordingly.
User Enrollment:
The program includes functions (new_student, new_employee, new_visitor) to enroll new users by appending their instances to the 
respective user lists. The functions also associate two known face images with each user and store the paths and names in the appropriate lists.

Program Execution:
The program starts by initializing the Student, Employee, and Visitor objects for existing users. It then enters the main function 
to handle access control based on face recognition. The program continuously loops, capturing faces, identifying users, and providing 
audio feedback until interrupted.

Time and Accuracy: 
To determine the accuracy of our face recgnition model, we carried out 100 tests, on 5 different humans, each human being subjected to 20 tests. On an 
average, the model delivered successful recognition 15.82 out of 20 times. In an ideal situation, the time taken by the model for 1 iteration was 
measured to be approximately 3 seconds for the facial recognition and 8 seconds for the complete interaction. 

Conclusion:
The Face Recognition-based Campus Access System demonstrates the use of face recognition technology to control access to campus areas. 
By combining face detection, recognition, and user management, the program enables seamless identification and access control for students, 
employees, and visitors. It provides audio feedback and prompts users for voice input, enhancing the user experience and ensuring a secure and 
convenient campus access system.

**************************************************************************************************************************************************