Automated border control systems are the first critical infrastructure point when crossing a border country. Crossing border lines for unauthorized passengers is a high security risk to any country. This paper presents a multispectral analysis of presentation attack detection for facial biometrics using the learned features from a convolutional neural network. Three sensors are considered to design and develop a new database that is composed of visible (VIS), near-infrared (NIR), and thermal images. Most studies are based on laboratory or ideal conditions-controlled environments. However, in a real scenario, a subject’s situation is completely modified due to diverse physiological conditions, such as stress, temperature changes, sweating, and increased blood pressure. For this reason, the added value of this study is that this database was acquired in situ. The attacks considered were printed, masked, and displayed images. In addition, five classifiers were used to detect the presentation attack. Note that thermal sensors provide better performance than other solutions. The results present better outputs when all sensors are used together, regardless of whether classifier or feature-level fusion is considered. Finally, classifiers such as KNN or SVM show high performance and low computational level.

Presentation attacks are one of the many dangers facing law enforcement today. In addition, material science is constantly advancing and criminals, aware of this fact, are taking advantage of new composites to manufacture new artifacts that allow them to cross borders by breaching border control points. This article presents the creation of several presentation attacks using make-up, hyper-realistic latex, and prosthetic masks. It is worth noting that such attacks do not receive adequate attention, due to the difficulty in their elaboration. The work of professionals in the make-up sector is required. Each stage of processing is analyzed for any artifacts that would facilitate the detection of the attack, using a multispectral approach in the visible and thermal spectra. The methodology evaluates three different face recognition systems (FRS), the different stages of impersonation, i.e. when a specific part of the face such as nose, cheekbones, jaw, or eyes is incorporated. The results show that certain parts of the face improve impersonation and make it more difficult for the algorithms to detect possible impersonation. However, other parts of the face, such as the jaw, not only do not improve impersonation but also significantly worsen performance. Using OpenFace as an FRS example, which is one of the FRS employed in this research work, the bonafide comparison of the target yields a score of 0.304, while with the make-up attack before applying make-up to the jaw, it gives 0.291, and after applying make-up, it gives 0.421.

This study introduces a reinforcement training framework for face recognition systems (FRSs) that leverages facial morphing techniques to generate counterfactual visual instances for model enhancement. Two complementary morphing strategies were employed: a geometric approach based on Delaunay–Voronoi triangulation (DVT‐Morph) and a generative approach using latent diffusion and autoencoder‐based models (diffusion‐based morphing [MorDIFF]). The generated morphs act as controlled counterfactuals, representing minimally modified facial images that induce changes in FRS verification decisions. The proposed method integrates these counterfactuals into the training process of two state‐of‐the‐art recognition systems, ArcFace and MagFace, to strengthen their decision boundaries and improve their robustness, calibration, and explainability. By combining morphing‐based counterfactual generation with eXplainable Artificial Intelligence (XAI) techniques, the framework enables a more interpretable embedding space and increased resilience against morphing and adversarial perturbations. The experimental results demonstrate that the inclusion of morph‐based counterfactuals significantly enhances the verification accuracy and decision transparency of modern FRSs. Moreover, the methodology is model‐ and morphing‐agnostic and can be applied to any FRS architecture, regardless of the morphing generation technique.

This paper proposes a theoretical framework that defines a generic and universal structure for Final Degree Projects in the area of Information and Communication Technologies that aims to be a formative element in itself, offering students a structured approach to successfully complete their projects, applying and developing their computational skills and information literacy competencies acquired during their formal university education.