ABSTARCT :
Current face biometric systems are vulnerable to spoofing attacks. A spoofing attack occurs when a person tries to masquerade as someone else by falsifying data and thereby gaining illegitimate access.
Inspired by image quality assessment, characterization of printing artefacts, and differences in light reflection, we propose to approach the problem of spoofing detection from texture analysis point of view.
Indeed, face prints usually contain printing quality defects that can be well detected using texture and local shape features. Hence, we present a novel approach based on analyzing facial image for detecting whether there is a live person in front of the camera or a face print.
The proposed approach analyzes the texture and gradient structures of the facial images using a set of low-level feature descriptors, fast linear classification scheme and score level fusion. Compared to many previous works, our proposed approach is robust and does not require user-cooperation.
In addition, the texture features that are used for spoofing detection can also be used for face recognition. This provides a unique feature space for coupling spoofing detection and face recognition. Extensive experimental analysis on three publicly available databases showed excellent results compared to existing works.
EXISTING SYSTEM :
In this work, we extend our spoofing detection approach using local binary pattern (LBP) based micro-texture analysis [7] by introducing two low-level features, Gabor wavelet features [8] and Histogram of Oriented Gradients (HOG) [9], to the face description which now consists of three enhanced feature vectors.
The proposed method adopts complementary properties of two powerful texture descriptors, since LBP encodes the micro-texture patterns and Gabor filters more macroscopic information.
In addition, HOG based local shape description provides additional information to the face description. A homogeneous kernel map [10] is applied on each resulting feature vector transforming the data into compact linear representation and reproducing an accurate approximation of the desired kernel function.
This representation enables then to use fast linear support vector machine (SVM) [11] classifiers. The final decision, whether there is a live person in front of the camera or not, is based on the score level fusion of the individual SVM outputs.
DISADVANTAGE :
Vulnerability to Advanced Spoofing Techniques :Sophisticated spoofing methods, such as 3D face masks or deepfake technology, can sometimes bypass detection systems, making them less effective against new types of spoofing attacks.
Hardware Requirements :High-quality face spoofing detection often requires advanced hardware, such as infrared cameras, structured light, or depth sensors, which can increase the cost and complexity of the system.
Privacy Concerns :Continuous facial recognition and spoofing detection could lead to privacy concerns as users’ biometric data is being collected and processed, potentially increasing the risk of misuse or unauthorized data access.
Environmental Factors :Lighting conditions, camera angle, and other environmental factors can affect the accuracy of face spoofing detection systems, leading to either poor detection or false rejection.
PROPOSED SYSTEM :
Spoofing occurs when the attacker presents a non-real image or sample of identity of valid user to the acquisition sensor.
In this proposed system there are two stages:
Face detection in an image
ii. Face Verification of real or spoof face in an image,
ADVANTAGE :
Prevention of Fraudulent Access: Face spoofing detection prevents unauthorized access to sensitive systems by detecting attempts to impersonate users with photos, videos, or masks.
Countering Sophisticated Spoofing Techniques: Modern spoofing methods, such as 3D masks, deepfake videos, or printed photos, can be effectively detected using advanced spoofing detection methods, ensuring that even the latest fraud techniques are thwarted.
Reliable Biometric Identification: By verifying that the face being scanned is live and authentic, spoofing detection increases the reliability and trustworthiness of biometric authentication.
Efficient and Touchless Access: Face recognition systems with spoofing detection provide fast, touchless, and frictionless authentication, improving user experience compared to other methods like PIN entry or fingerprint scanning.
|