Extended StirTrace benchmarking of biometric and forensic qualities of morphed face images

Abstract

Since its introduction in 2014, the face morphing forgery (FMF) attack has received significant attention from the biometric and media forensic research communities. The attack aims at creating artificially weakened templates which can be successfully matched against multiple persons. If successful, the attack has an immense impact on many biometric authentication scenarios including the application of electronic machine-readable travel document (eMRTD) at automated border control gates. We extend the StirTrace framework for benchmarking FMF attacks by adding five issues: a novel threefold definition for the quality of morphed images, a novel FMF realisation (combined morphing), a post-processing operation to simulate the digital image format used in eMRTD (passport scaling 15 kB), an automated face recognition system (VGG face descriptor) as additional means for biometric quality assessment and two feature spaces for FMF detection (keypoint features and fusion of keypoint and Benford features) as additional means for forensic quality assessment. We show that the impact of StirTrace post-processing operations on the biometric quality of morphed face images is negligible except for two noise operators and passport scaling 15 kB, the impact on the forensic quality depends on the type of post-processing, and the new FMF realisation outperforms the previously considered ones.

Computer Science & Information Technology (CS & IT)

“…The second validation set we use is the "AMSL Face Morph Image Data Set" dataset introduced in [23]. These morphs were created using images from [24] and [25].…”

A Face Morphing Detection Concept with a Frequency and a Spatial Domain Feature Space for Images on eMRTD

“…For our paper, we use three state-of-the-art morph generation pipelines (MGP) to create three dierent types of face morphing for our evalatuion dats sets from two related works [19,24]. As mentioned in [30], all morph pipelines are based on alpha-blending and warping.…”

“…• The combined morph is the third MGP used and combines the aforementioned pipelines to avoid shortcomings. The pipeline is introduced in [24] and aligns the original images prior to the warping process to ensure that warping does not lead to a distortion of the face geometry. This MGP has the best visual quality of the three mentioned pipelines as well as a high biometric quality [24].…”

“…The work in [24] shows that this detector is not suitable for passportscaled images, because the reduction of the lesize and the downscaling also lead to a reduction of keypoints and edges. So, the detector delivers a reasonable high false alarm rate (ca.…”

Novel framework for optimized digital forensic for mitigating complex image attacks

“…Reillo et al [15] have used attribute-based identification of varied signatures of humans by extracting the geometric attributes. The work of Neubert et al [16] has presented a solution towards resisting forging face shape based attack using a morphing based approach for better biometric security. Guo et al [17] have carried out an identification of the image based on forged color information using histogram-based as well as an attribute-based encoding mechanism.…”