Human face identification via comparative soft biometrics

Relevance. Nowadays, technological systems, artificial intelligence, the general availability of the Internet and penetration into the systems of banks, institutions and social networks have become a studied science and are accessible to all groups and ages. One of the main tasks was to provide a system for protecting confidential information from hackers, as well as easy access to authentication and identification of users. Biometric systems came to the fore, including mouse movement dynamics and keystroke dynamics, which reveal the typing style and mouse movement of each person. Soft biometrics is an interesting and inexpensive biometric method that does not require additional equipment. The system identifies a person based on the input information they enter in a special column. Hand identification dynamics falls into the category of behavioral soft biometrics, that is, the user's patterns reflect the individual program of actions that he follows when using the site.The goal of this article the purpose of this work is to improve the security level by creating a function that will strengthen the authentication system and improve the iron gate Методы исследования. In carrying out the work, methods of analysis and synthesis, theories of algorithms, laws of kinematics, neural networks, keystroke dynamics and soft biometrics were used.Results. A method for extracting dynamic characteristics of keystrokes is described. A neural network is created and a threshold value is determined for identifying the type of typing hand.Scientific novelty. Unlike known authentication methods, the proposed method is used to determine the typing hand on the keyboard through a neural network using the laws of kinematics, soft biometrics and extracting the dynamics of keystrokes in order to determine the value and accuracy of determining the type of typing hand.Significance. The proposed solution allows to increase the security of user authentication, increase the speed of implementation and reduce the cost. The results obtained in the work are positive and can be used in the near future. In turn, soft biometric measurements depend on human behavioral patterns, which complicates user falsification. It is difficult to imitate typing behavior, since it is ballistic (semi-autonomous), which makes behavioral information valuable as a soft and sensitive biometric method.

Matching people in different camera views, commonly referred to as person re-identification, is an inherently challenging task due to the appearance disparity caused by view change.Other factors such as low image resolution and occlusion further compound this problem. As a highly demanded technique, person re-identification has been actively studied in recent years. Most of the existing approaches either focus on feature design or distance metric learning, based on appearance features. However, due to the view change, the appearance features may significantly vary for the same subject, resulting in matching difficulties. Instead of using features from a single modality, i.e, appearance, we propose to use multimodal features to improve the re-identification accuracy. Specifically, in this work, we leverage both appearance features and soft biometrics, i.e, human characteristics such as gender, to match individuals across cameras. We build multiple graphs, each of which represent one feature modality, and the graphs are then combined and optimized to derive the similarities between a probe and the gallery subjects. The proposed method is evaluated on the VIPeR dataset with annotated soft biometric labels. The results suggest that using multimodal features, e.g, appearance and soft biometrics, can improve the matching accuracy as compared to using appearance features only, and superior performance is obtained as compared to other state-of-the-art approaches.

The proliferate unstructured data generated in online social networks leads to significant research advances in the recognition of user profiles (e.g., age, gender, ethnicity, etc.), but meanwhile brings new challenges. These attributes are referred to as soft biometrics and provide a semantic description of users. Identifying users' soft biometric traits in online social networks is crucial for a variety of applications such as customized marketing, personalized recommendation, and urban planning. Compared to conventional studies, we address the importance and challenges of identifying soft biometrics in online social networks and provide a case study on gender recognition of Twitter users based on their tweets texts and profile images. We first apply an efficient approach to label the soft biometric attributes of users using their self-reported names, instead of labor-forced methodology such as manually labeling and crowdsourcing. Then we investigate the approaches using texts and profile images, respectively, including Hashtag-based TF-IDF, LDA Topic Distribution, SIFT-based TF-IDF and Convolutional Neural Networks (CNN). Finally, we propose an ensemble method by combining models learned from both texts and images to enhance gender identification. The performance in our extensive experiments is demonstrated to be comparative with state-of the-art work.

This paper presents a template aging study of eye movement biometrics, considering three distinct biometric techniques on multiple stimuli and eye tracking systems. Short-to-midterm aging effects are examined over two-weeks, on a highresolution eye tracking system, and seven-months, on a low-resolution eye tracking system. We find that, in all cases, aging effects are evident as early as two weeks after initial template collection, with an average 28% (±19%) increase in equal error rates and 34% (±12%) reduction in rank-1 identification rates. At seven months, we observe an average 18% (±8%) increase in equal error rates and 44% (±20%) reduction in rank-1 identification rates. The comparative results at two-weeks and seven-months suggests that there is little difference in aging effects between the two intervals; however, whether the rate of decay increases more drastically in the long-term remains to be seen.

An important challenge in the research field of Biometrics isreal-time identification, at a distance, in uncontrolled environments, us-ing low-resolution cameras. In such circumstances, soft biometrics can bethe only option. In this work, we propose two novel descriptor methodsfor biometric identification based on ensemble of anthropometric mea-surements and joints heat-map of the person skeleton, captured fromvideo frames through state-of-the-art 2D poses estimation methods. Theproposed methods were assessed on a popular benchmark dataset, CA-SIA Gait Dataset B, and obtained good results (85% and 89% of rank-1identification rates, respectively) with PifPaf 2D pose estimation method

Person re-identification plays an important role in the construction of the smart city. A reliable person re-identification system relieves users from the inefficient work of identifying the specific individual from enormous numbers of photos or videos captured by different surveillance devices. The most existing methods either focus on local discriminative features without global contextual information or scatter global features while ignoring the local features, resulting in ineffective attention to irregular pedestrian zones. In this article, a novel Transformer-CNN Coupling Network (TCCNet) is proposed to capture the fluctuant body region features in a heterogeneous feature-aware manner. We employ two bridging modules, the Low-level Feature Coupling Module (LFCM) and the High-level Feature Coupling Module (HFCM), to improve the complementary characteristics of the hybrid network. It is significantly helpful to enhance the capacity to distinguish between foreground and background features, thereby reducing the unfavorable impact of cluttered backgrounds on person re-identification. Furthermore, the duplicate loss for the two branches is employed to incorporate semantic information from distant preferences of the two branches into the resulting person representation. The experiments on two large-scale person re-identification benchmarks demonstrate that the proposed TCCNet achieves competitive results compared with several state-of-the-art approaches. The mean Average Precision (mAP) and Rank-1 identification rate on the MSMT17 dataset achieve 66.9% and 84.5%, respectively.

Annotating age classes for humans’ facial images according to their appearance is very challenging because of dynamic person-specific ageing pattern, and thus leads to a set of unreliable apparent age labels for each image. For utilising ambiguous label annotations, an intuitive strategy is to generate a pseudo age for each image, typically the average value of manually-annotated age annotations, which is thus fed into standard supervised learning frameworks designed for chronological age estimation. Alternatively, inspired by the recent success of label distribution learning, this paper introduces a novel concept of ambiguous label distribution for apparent age estimation, which is developed under the following observations that (1) soft labelling is beneficial for alleviating the suffering of inaccurate annotations and (2) more reliable annotations should contribute more. To achieve the goal, label distributions of sparse age annotations for each image are weighted according to their reliableness and then combined to construct an ambiguous label distribution. In the light, the proposed learning framework not only inherits the advantages from conventional learning with label distribution to capture latent label correlation but also exploits annotation reliableness to improve the robustness against inconsistent age annotations. Experimental evaluation on the FG-NET age estimation benchmark verifies its effectiveness and superior performance over the state-of-the-art frameworks for apparent age estimation.

Recently, multilabel classification algorithms play an increasingly significant role in data mining and machine learning. However, some existing mutual information-based algorithms ignore the influence of the proportions of labels on the correlation degree between features and label sets. Besides, the correlation degree of label sets cannot be accurately measured in most traditional ReliefF algorithms, and the repeated calculation arises from the division of heterogeneous neighbors. To overcome these shortcomings, this paper proposes a multilabel feature selection method using mutual information and improved multilabel ReliefF (ML-ReliefF). First, the proportion of each label is calculated in label space and combined with the mutual information of features and labels to construct a novel correlation degree between features and label sets to preprocess multilabel datasets, which is used to reduce runtime of ML-ReliefF. Second, the mutual information of label sets is introduced into improving accuracy of the correlation degree among label sets. Furthermore, two types of correlation degree for label sets based on ML-ReliefF are developed to divide similar and heterogeneous samples more clearly. Third, a divided method of heterogeneous neighbors is presented to effectively avoid the repeated calculation in ML-ReliefF, and a novel method of feature weighting based on ML-ReliefF is constructed to evaluate the importance of features. Finally, a multilabel feature selection algorithm based on mutual information and ML-ReliefF for multilabel classification is designed to improve the performance of multilabel classification. Experiments under fourteen multilabel datasets show the effectiveness of our algorithm and improve the classification performance for multilabel datasets.

Feature selection has attracted considerable attention due to the wide application of multi-label learning. However, previous methods do not fully consider the relationship between feature sets and label sets but devote attention to either of them. Furthermore, conventional multi-label learning utilizes logical labels to estimate relevance between feature sets and label sets so that the importance of corresponding labels cannot be well reflected. Additionally, numerous irrelevant and redundant labels degrade the classification performance of models. To this end, we propose a multi-label feature selection method named Robust multi-label Feature Selection with shared Label Enhanced (RLEFS). First, we obtain a robust label enhancement term by reconstructing labels from logical labels to numerical labels and imposing \(l_{2,1}\)-norm onto the label enhancement term. Second, RLEFS utilizes the robust label enhancement term to share the similar latent semantic structure between feature matrix and label matrix. Third, local structure is considered to ensure the consistency of label information during the feature selection process. Finally, we integrate the above terms into one joint learning framework, and then, a simple but effective optimization method with provable convergence is proposed to solve RLEFS. Experimental results demonstrate the classification superiority of RLEFS in comparison with seven state-of-the-art methods.

A label-specific multi-label feature selection algorithm based on the Pareto dominance concept

This study compares the performance of XGBoost and LightGBM models for predicting live weights of Yucatecan Criollo pigs from biometric measurements and examines the structural and algorithmic differences that affect model fit. Detailed analysis of the models' hyperparameter optimization and variable importance revealed how each model approaches the data and prioritizes features. This study was conducted on 182 Yucatecan Criollo pigs (134 females and 48 males). When model performances were evaluated, the XGBoost model showed superior prediction performance with acceptable accuracy and lower error rates in the test dataset, with R2 = 0.905, RMSE = 5.704, and MAE = 3.636. In contrast, the LightGBM model produced acceptable results under certain hyperparameter combinations with R2 = 0.824, RMSE = 7.772, and MAE = 5.505. While the robust performance of both models requires strategic decisions in model selection and optimization, it is recommended to consider the dataset's nature in feature selection and hyperparameter settings. This study provides important insights for simplifying the model and improving its efficiency in machine learning applications, and serves as a reference for more effective model use.

Robust feature selection using label enhancement and β-precision fuzzy rough sets for multilabel fuzzy decision system

Slot filling is one of the critical tasks in modern conversational systems. The majority of existing literature employs supervised learning methods, which require labeled training data for each new domain. Zero-shot learning and weak supervision approaches, among others, have shown promise as alternatives to manual labeling. Nonetheless, these learning paradigms are significantly inferior to supervised learning approaches in terms of performance. To minimize this performance gap and demonstrate the possibility of open-domain slot filling, we propose a Self-supervised Co-training framework, called , that requires zero in-domain manually labeled training examples and works in three phases. Phase one acquires two sets of complementary pseudo labels automatically. Phase two leverages the power of the pre-trained language model BERT, by adapting it for the slot filling task using these sets of pseudo labels. In phase three, we introduce a self-supervised co-training mechanism, where both models automatically select high-confidence soft labels to further improve the performance of the other in an iterative fashion. Our thorough evaluations show that outperforms state-of-the-art models by 45.57% and 37.56% on SGD and MultiWoZ datasets, respectively. Moreover, our proposed framework achieves comparable performance when compared to state-of-the-art fully supervised models.

Feature selection can remove non-important features from the data and promote better classifiers. This task, when applied to multi-label data where each instance is associated with a set of labels, supports emerging applications. Although multi-label data usually exhibit label relations, label dependence has been little studied in feature selection. We proposed two multi-label feature selection algorithms that consider label relations. These methods were experimentally competitive with traditional approaches. Moreover, this work conducted a systematic literature review, summarizing 74 related papers.

Multi-label classification task has many applications in Text Categorization, Multimedia, Biology, Chemical data analysis and Social Network Mining, among others. Different approaches have been developed: Binary Relevance (BR), Label Power Set (LPS), Random k label sets (RAkEL), some of them consider the interaction between labels in a chain (Chain Classifier) and other alternatives around this method are derived, for instance, Probabilistic Chain Classifier, Monte Carlo Chain Classifier and Bayesian Chain Classifier (BCC). All previous approaches have in common and focus on is in considering different orders or combinations of the way the labels have to be predicted. Given that feature selection has proved to be important in classification tasks, reducing the dimensionality of the problem and even improving classification model’s accuracy. In this work a feature selection technique is tested in BCC algorithm with two searching methods, one using Best First (BF-FS-BCC) and another with GreedyStepwise (GS-FS-BCC), these methods are compared, the winner is also compared with BCC, both tests are compared through Wilcoxon Signed Rank test, in addition it is compared with others Chain Classifier and finally it is compared with others approaches (BR, RAkEL, LPS).