Failure of facial recognition and authentication system may lead to several unlawful activities. The current facial recognition systems are vulnerable to different biometric attacks. This research focuses on morphing attack detection. This research proposes a robust detection mechanism that can deal with variation in age, illumination, eye and head gears. A deep learning based feature extractor along with a classifier is adopted. Additionally, image enhancement and feature combination are proposed to augment the detection results. A versatile dataset is also developed that contains Morph-2 and Morph-3 images, created by sophisticated tools with manual intervention. Morph-3 images can give more realistic appearance and hence difficult to detect. Moreover, Morph-3 images are not considered in the literature before. Professional morphing software depicts more realistic morph attack scenario as compared to the morphs generated in the previous work from free programs and code scripts. Eight face databases are used for creation of morphs to encompass the variation. These databases are Celebrity2000, Extended Yale, FEI, FGNET, GT-DB, MULTI-PIE, FERET and FRLL. Results are investigated using multiple experimental setups and it is concluded that the proposed methodology gives promising results.

A robust facial recognition system that has soundness and completeness is essential for authorized control access to lawful resources. Due to the availability of modern image manipulation technology, the current facial recognition systems are vulnerable to different biometric attacks. Image morphing attack is one of these attacks. This paper compares and analyzes state-of-the-art morphing attack detection (MAD) methods. The performance of different MAD methods is also compared on a wide range of source image databases. Moreover, it also describes the morph image generation techniques along with the limitations, strengths, and drawbacks of each morphing technique. Results are investigated and compared with in-depth analysis providing insight into the vulnerabilities of existing systems. This paper provides vital information that is essential for building a next generation morph attack detection system.