Face Recognition Research Articles

Development of face image recognition algorithm using CNN in airport security checkpoints for terrorist early detection

Ensuring airport security is of paramount importance to safeguard the lives of passengers and prevent acts of terrorism. In this context, developing advanced technology for early terrorist detection is crucial. This paper presents a novel approach to enhancing security measures at airport checkpoints by applying Convolutional Neural Network (CNN) and Artificial Neural Network (ANN) algorithms in face image recognition. Our system utilizes state-of-the-art artificial intelligence techniques to analyze facial features. Our research uses VGG architecture and pre-trained with face data as a CNN model. This model is used to extract face embedding features from the dataset. These embedding features are then compressed with Principal Component Analysis (PCA) to obtain the meaningful feature as training data for the ANN algorithm. We trained our system using data from 500 identities data with 60 data for each identity. This training enables our system to recognize known terrorists and individuals on watchlists by comparing the facial features of individuals passing through security checkpoints with those in the database. The proposed CNN-ANN-based face recognition system not only enhances airport security but also significantly reduces the processing time for security checks. It can quickly identify potential threats, allowing security personnel to take appropriate actions in real time ensuring a rapid response to security concerns. We present the architecture, training methodology, and evaluation of the CNN-ANN model, achieving a high accuracy of 91.16% and precision of 91.36%. Through this research, we aim to increase airport security and strengthen efforts to combat terrorism, making air travel safer and more secure for all passengers.

Research and Analysis of Facial Recognition Based on FaceNet, DeepFace, and OpenFace

This study provides a comprehensive review of recent advancements in face recognition technology, focusing on deep learning models such as FaceNet, DeepFace, and OpenFace. The primary evaluation criterion is these models' ability to produce accurate facial embeddings, which are essential for reliable identification and verification. The findings demonstrate that these models significantly enhance recognition performance, particularly under challenging conditions such as varying lighting and occlusions. However, the study also identifies ongoing issues, including the need for efficient processing and reliance on large, annotated datasets. Future research should address these challenges by improving the efficiency and scalability of deep learning models. Additionally, expanding datasets to include a broader range of facial features will enhance model robustness in real-world applications. Exploring the integration of advanced technologies, such as sophisticated data augmentation techniques, will further boost the accuracy and adaptability of face recognition systems. These efforts are expected to advance the development of more versatile and reliable face recognition technologies.

Facial emotion recognition based on upper features and transfer learning

Facial expression recognition (FER) in the upper face focuses on the analysis and recognition of emotions based on features extracted from the upper region of the face. This region typically affects the eyes, eyebrows, forehead, and sometimes the upper cheeks. Since these areas are often less affected by face masks or other facial coverings, FER algorithms can concentrate on capturing and interpreting the relevant facial cues, such as eye movements, eyebrow positions, and forehead wrinkles, to accurately recognize and classify different emotions. By focusing on the upper face, FER systems can mitigate the impact of occlusions caused by masks and still provide meaningful insights into the emotional states of individuals. In this work, a FER approach focusing on the upper region is proposed. Several experiments have been made using the CK+ dataset in addition to a comparison between the emotion recognition scores using the upper and the entire face in order to determine whether this area can reflect the real expressed emotion. The results of our approach are promising compared to previous studies with an accuracy up to 96%.

Face Recognition and Brand Strategy: The Market Competitiveness of Face Graphic Logos

Face Recognition and Brand Strategy: The Market Competitiveness of Face Graphic Logos

Multimodal deep learning-based feature fusion for object detection in remote sensing images

Object detection is an important computer vision task, which is developed from image classification task. The difference is that it is no longer only to classify a single type of object in an image, but to complete the classification and positioning of multiple objects that may exist in an image at the same time. Classification refers to assigning category labels to the object, and positioning refers to determining the vertex coordinates of the peripheral rectangular box of the object. Therefore, object detection is more challenging and has broader application prospects, such as automatic driving, face recognition, pedestrian detection, medical detection etc,. Object detection can also be used as the research basis for more complex computer vision task such as image segmentation, image description, object tracking and action recognition. In traditional object detection, the feature utilization rate is low and it is easy to be affected by other environmental factors. Hence, this paper proposes a multimodal deep learning-based feature fusion for object detection in remote sensing images. In the new model, cascade RCNN is the backbone network. Parallel cascade RCNN network is utilized for feature fusion to enhance feature expression ability. In order to solve the problem of different segmentation shapes and sizes, the central part of the network adopts multi-coefficient cascaded hollow convolution to obtain multi-receptive field features without using pooling mode and preserving image information. Meanwhile, an improved self36 attention combined receptive field strategy is used to obtain both low-level features with marginal details and high-level features with global semantics. Finally, we conduct experiments on DOTA set including ablation experiments and comparison experiments. The experimental results show that the mean Average Precision (mAP) and other indexes have been greatly improved, and its performance is better than the state-of-the-art detection algorithms. It has a good application prospect in the remote sensing image object detection task.

Generalized Relevance Learning Grassmann Quantization.

Due to advancements in digital cameras, it is easy to gather multiple images (or videos) from an object under different conditions. Therefore, image-set classification has attracted more attention, and different solutions were proposed to model them. A popular way to model image sets is subspaces, which form a manifold called the Grassmann manifold. In this contribution, we extend the application of Generalized Relevance Learning Vector Quantization to deal with Grassmann manifold. The proposed model returns a set of prototype subspaces and a relevance vector. While prototypes model typical behaviours within classes, the relevance factors specify the most discriminative principal vectors (or images) for the classification task. They both provide insights into the model's decisions by highlighting influential images and pixels for predictions. Moreover, due to learning prototypes, the model complexity of the new method during inference is independent of dataset size, unlike previous works. We applied it to several recognition tasks including handwritten digit recognition, face recognition, activity recognition, and object recognition. Experiments demonstrate that it outperforms previous works with lower complexity and can successfully model the variation, such as handwritten style or lighting conditions. Moreover, the presence of relevances makes the model robust to the selection of subspaces' dimensionality.

How the brain differentiates human and monkey faces: Neuromagnetic evidence from spatiotemporal dynamics

Face recognition is an important aspect of human non-verbal communication. Event-related potentials or magnetic fields, such as the N170/M170 component, are considered essential neural markers of face processing. Compared to upright human faces, inverted human faces and upright but not inverted animal faces cause longer latencies and larger amplitudes of these components. However, the mechanisms underlying this factor remain unclear. To elucidate the spatiotemporal dynamics of the processing of inverted human and animal faces, we recorded face-selective responses (M170) to upright and inverted human and monkey faces using a 306-channel whole-head magnetoencephalography. Sensor analysis showed an increased M170 latency and amplitude for inverted human and upright animal faces. However, in the source analysis, the observed modulations of the estimated spatiotemporal dynamics were different from the sensor results: irrespective of species, upright faces activated wider areas in the ventral and dorsal visual regions compared with inverted faces. Additionally, face orientation differentially modulated the anterior region of the fusiform gyrus (FG) in both face categories. These results suggest that spatiotemporal dynamics differ in face orientation regardless of category and that the FG contributes little or nothing to the M170 modulation recorded in the scalp sensor. Furthermore, we demonstrated that inverted human and animal faces are processed via different mechanisms.

A Graph-Based Statistical Approach to Identifying Functional Connectivity Networks in Patients with Traumatic Brain Injury.

Traumatic brain injury (TBI) is one of the most common types of brain injuries associated with cognitive impairments. Functional magnetic resonance imaging (fMRI) studies can provide a unique opportunity to examine brain connectivity patterns and understand the neural substrates of cognitive outcomes following traumatic injury. Therefore, this study aims to determine changes in functional connectivity patterns in patients with TBI compared to healthy individuals using two graph models, adaptive dense subgraph discovery (ADSD) and variance component. This study used fMRI data downloaded from https://openneuro.org. These data included 14 patients with TBI aged between 18 and 36 and 12 healthy individuals (female: N=6, male: N=6) aged between 19 and 52. Out of the 74 regions examined, a cluster of 18 regions related to TBI was identified using the ADSD model. Subsequently, these identified regions were used as input for the variance component model to investigate changes in connectivity patterns. Functional connectivity between an 18-brain region cluster, such as the Rectus (Left, Right), Supp_Motor_Area (Left, Right), and Middle Cingulum (Left, Right), differed between the patient and healthy groups. Based on the analysis of functional connectivity between pairs of brain regions, 153 connections between pairs of brain regions were compared in the two groups, out of which 63 connections showed significant differences between the two groups. Compared to other regions, Supp_Motor_Area_Right and Rectus_Left had more connections. The study's results indicate that the functional connectivity between the Cingulum, Hippocampus, Fusiform, Supp_Motor_Area, and Precentral regions differs between the two groups. Since these regions are involved in processes such as memory, learning, spatial orientation, face recognition, coordination, and motor control, changes in their functional connectivity may lead to impairments in these areas.

Analysis of The Role of Deep Learning Models in Image Classification Applications

Image classification is a fundamental task in computer science, underpinning various applications such as object detection, face recognition, and object interaction analysis. The concept holds significant value due to its wide-ranging applications across multiple fields. Traditional methods for image classification, however, have been limited by their slow processing speed, rigidity, and high costs. The integration of deep learning models, particularly Convolutional Neural Networks (CNNs) and Support Vector Machines (SVMs), has revolutionized this process, enabling the development of automated, fast, and practical systems. These advanced models are now employed in diverse areas, including biomedical science, remote sensing, and business management, thanks to their ability to achieve high accuracy across a broad spectrum of scenarios. Training these models involves the use of well-known datasets like Canadian Institute for Advanced Research (CIFAR) and Modified National Institute of Standards and Technology (MNIST), which provide the necessary data for optimization and validation. The paper examines the structure, functionality, advantages, and limitations of CNNs and SVMs in the context of image classification, demonstrating that deep learning-driven classification is now a mainstream research focus. This study highlights the transformative impact of these models and provides insights into their future potential.

Enhancing Handwritten Text Recognition on Mobile Platforms Using Cloudlet-Assisted Deep Learning

The limited computational resources of mobile devices significantly constrain their ability to effectively execute resource-intensive computer vision tasks, such as , video processing, augmented reality, and face recognition. Handwritten character recognition (HCR) is a critical application of machine learning and computer vision, with wide-ranging implications for digital transformation and accessibility. This proposed study explores the implementation of a deep learning-based HCR system tailored for mobile phones using cloudlet. A cloudlet is a novel technology aimed at boosting the computational power of mobile devices to manage resource-intensive tasks. It enables mobile clients to interact with nearby servers, allowing only minimally processed data to be transmitted from the mobile application to the server. The server handles the complex computations and sends the results back to the mobile device in real-time. By utilizing Convolutional Neural Networks (CNNs), the proposed method achieves high accuracy and efficiency on resource-constrained mobile devices. The research details the system architecture, dataset preprocessing, model training, and integration into an Android application, offering a robust solution for real-time handwritten character recognition.

The face recall and face recognition: description accuracy does not indicate eyewitness identification accuracy

Eyewitness testimonies are frequently considered crucial in criminal investigations. The impact of the initial description provided by an eyewitness on subsequent identification accuracy remains an understudied area. We employed a slightly modified version of Schooler and Engstler-Schooler's experimental approach to assess whether description quantity and quality are related to lineup identification accuracy. The sample comprised 99 undergraduate students from the University of Sarajevo. More than 50% of our participants made an accurate lineup identification, whereas they recalled and described only the general physical characteristics with minimal detail. No significant relationship between the quantity/quality of descriptions and identification accuracy was observed in terms of the total number of words, total number of details in the description, and individual characteristics that participants mentioned in the description. In other words, neither an extensive nor a detailed description effectively predicts lineup identification accuracy. Finally, our findings suggest that the recognition task is significantly less challenging for participants than the recall and description tasks because descriptions provide a general impression of an individual and lack detailed information. The findings of this study indicate that a more detailed and extensive description does not serve as a reliable indicator of lineup identification accuracy.

Two-step authentication: multi-biometric system using voice and facial recognition

Two-step authentication: multi-biometric system using voice and facial recognition

On Privacy, Accuracy, and Fairness Trade-offs in Facial Recognition

On Privacy, Accuracy, and Fairness Trade-offs in Facial Recognition

From Theory to Implementation: Optimizing AI-Driven Depression Detection Using Facial Recognition, EEG, and Algorithmic Innovations

From Theory to Implementation: Optimizing AI-Driven Depression Detection Using Facial Recognition, EEG, and Algorithmic Innovations

Neural Pathways to Emotional Wellness: Merging AI-Driven VPSYC Systems with EEG and Facial Recognition

Neural Pathways to Emotional Wellness: Merging AI-Driven VPSYC Systems with EEG and Facial Recognition

EmoSense: Pioneering Facial Emotion Recognition with Precision Through Model Optimization and Face Emotion Constraints

EmoSense: Pioneering Facial Emotion Recognition with Precision Through Model Optimization and Face Emotion Constraints

Facial Feature Recognition with Multi-task Learning and Attention-based Enhancements

Facial Feature Recognition with Multi-task Learning and Attention-based Enhancements

Emotions at the heart of learning: exploring the role of teacher emotions in student engagement using facial emotion recognition

Emotions at the heart of learning: exploring the role of teacher emotions in student engagement using facial emotion recognition

Scalable Context-Based Facial Emotion Recognition Using Facial Landmarks and Attention Mechanism

Scalable Context-Based Facial Emotion Recognition Using Facial Landmarks and Attention Mechanism

Face de‐morphing based on identity feature transfer

AbstractFace morphing attacks have emerged as a significant security threat, compromising the reliability of facial recognition systems. Despite extensive research on morphing detection, limited attention has been given to restoring accomplice face images, which is critical for forensic applications. This study aims to address this gap by proposing a novel face de‐morphing (FD) method based on identity feature transfer for restoring accomplice face images. The method encodes facial attribute and identity features separately and employs cross‐attention mechanisms to extract identity features from morphed faces relative to reference images. This process isolates and enhances the accomplice's identity features. Additionally, inverse linear interpolation is applied to transfer identity features to attribute features, further refining the restoration process. The enhanced identity features are then integrated with the StyleGAN generator to reconstruct high‐quality accomplice facial images. Experimental evaluations on two morphed face datasets demonstrate the effectiveness of the proposed approach, improving the average restoration accuracy by at least 5% compared with other methods. These findings highlight the potential of this approach for advancing forensic and security applications.