Dhananjay Sunil Menon, Rakesh B Krishnan, Shreya Chaudhary
Check out our final report here .
Youtube : https://www.youtube.com/watch?v=CxtbcyLmjx0
This project focusses on imparting advanced image recognition features in security footage to improve real-time alerts and overall safety in banks. In the past, security cameras were used to record footage of everything that goes on in the bank. With the advanced security software, security cameras can send real time alerts to notify emergency services in case of a robbery. The advanced security system recognizes faces of employees and estimates their pose. In case on a robbery, an alarm alert can be sent to the police station by registering a body gesture from the employee.
The main objectives of this project is to recognize faces of bank employees working at the bank, understand body language and sound alarm for a special body gesture and explore various algorithms for face recognition and body pose estimation.
With increasing technology in various aspects of security in banks, security cameras have always been the same. Extensive investigation must be done after a crime in order to gather evidence and leads in finding criminals. With smart vision technology, a lot of time can be saved with proactive alarm systems and criminal information can easily be extracted after the crime. Such security features can make robberies more difficult and can dissuade such crimes in the future.
With high-tech security and low crime rates, people of a community can be safe and have peace of mind. Banks can also be saved from any loss of money. In some countries, security cameras are present extensively. This makes it easier to nab criminals and dissuade people from doing anything immoral and wrong.
The advanced security system recognizes faces of employees and estimates their pose. In case on a robbery, an alarm alert can be sent to the police station by registering a body gesture from the employee. Face recognition is a technique that identifies or locates human faces in digital images. A typical example of face detection occurs when we take photographs through our smartphones, and it instantly detects faces in the picture. Face detection is different from Face recognition. Face detection detects merely the presence of faces in an image while facial recognition involves identifying whose face it is. Face Recognition has many applications in biometrics, surveillance systems, information security and law enforcement.
With increasing technology in various aspects of security in banks, security cameras have always been the same. Extensive investigation must be done after a crime in order to gather evidence and leads in finding criminals.
With smart vision technology, a lot of time can be saved with proactive alarm systems and criminal information can easily be extracted after the crime. Such security features can make robberies more difficult and can dissuade such crimes in the future.
With high-tech security and low crime rates, people of a community can be safe and have peace of mind. Banks can also be saved from any loss of money.
In some countries, security cameras are present extensively. This makes it easier to nab criminals and dissuade people from doing anything immoral and wrong.
Cameras are present on many public places to improve security and this works fine with motion detectors if it is not necessary to identify the person on the picture automatically. This project focusses on imparting advanced object, face and gesture recognition features in security footage to improve real-time alerts and overall safety in banks. In the past, security cameras were used to record footage of everything that goes on in the bank. With the advanced security software, security cameras can send real time alerts to notify emergency services in case of a robbery. Once a face is detected in images they can be identified or matched with a given face database.
Incase face is detected to be employees’, the proactive alert will sound when there the pose is shown for more than 5 seconds in front of the camera.
Incase someone other than the employee tries to activate the alarm, then the alarm does not activate as the face of that person is not recognized.
People who enter the frame have their screenshots saved along with the time stamp.
Each picture is saved with the time stamp first.
Weapons are detected in real time using the Yolo algorithm.
This project is an advanced security system camera software with functionalities such as proactive security system alert, weapon detection and identity profile retrieval. The proactive system alert detects if a robbery is taking place by scanning the pose of the employees and sends a notification to emergency services. Weapon detection uses object detection to find a weapon. Identity profile retrieval is used to get information on people if an incident has taken place.
With increasing technology in various aspects of security in banks, security cameras have always been the same. Extensive investigation must be done after a crime to gather evidence and leads in finding criminals. With smart vision technology, a lot of time can be saved with proactive alarm systems and criminal information can easily be extracted after the crime. Such security features can make robberies more difficult and can dissuade such crimes in the future.
With high-tech security and low crime rates, people of a community can be safe and have peace of mind. Banks can also be saved from any loss of money. In some countries, security cameras are present extensively. This makes it easier to nab criminals and dissuade people from doing anything immoral and wrong.
This application can be extended to banks for better camera security. The required cameras can be installed at specific angles and locations in order to get the best view needed for the implementation of this application.