Recognition Applications Research Articles

Trusted emotion recognition based on multiple signals captured from video and its application in intelligent education

<abstract> <p>The emotional variation can reflect shifts in mental and emotional states. It plays an important role in the field of intelligent education. Emotion recognition can be used as cues for teachers to evaluate the learning state, analyze learning motivation, interest, and efficiency. Although research on emotion recognition has been ongoing for a long time, there has been a restricted emphasis on analyzing the credibility of the recognized emotions. In this paper, the origin, development, and application of emotion recognition were introduced. Then, multiple signals captured from video that could reflect emotion changes were described in detail and their advantages and disadvantages were discussed. Moreover, a comprehensive summary of the pertinent applications and research endeavors of emotion recognition technology in the field of education was provided. Last, the trend of emotion recognition in the field of education was given.</p> </abstract>

Enhancing the application of signal light recognition for the YOLOv8 model in complex traffic scenarios

Enhancing the application of signal light recognition for the YOLOv8 model in complex traffic scenarios

Combination of Lv-3DCNN Algorithm in Random Noise Environment and Its Application in Aerobic Gymnastics Action Recognition

Combination of Lv-3DCNN Algorithm in Random Noise Environment and Its Application in Aerobic Gymnastics Action Recognition

Application of Multimodal Feature Selection-Based Scene Recognition for Medical Education

Application of Multimodal Feature Selection-Based Scene Recognition for Medical Education

ZJUT-EIFD: A Synchronously Collected External and Internal Fingerprint Database.

External fingerprints (EFs) based only on epidermal information are vulnerable to spoofing attacks and non-ideal skin conditions. To solve such shortcomings, internal fingerprints (IFs) collected using optical coherence tomography (OCT) have been proposed and widely researched. However, the development of IF is limited by the lack of in-depth researches on the IF and the EF-IF interoperability, which is partially caused by the lack of public OCT database. The obvious gap in the applications of EF and IF recognition motivated us to design and publish a comprehensive fingerprint database containing both traditional EFs and OCT IFs, denoted as ZJUT-EIFD. To the best of our knowledge, ZJUT-EIFD is the first public database that combines OCT and total internal reflection (TIR) via synchronous acquisition, with 399 different fingers from 60 subjects. In this article, the composition of the database, the quality of EFs and IFs, and the verification performance of different types of fingerprints were detailed. In addition, potential application directions of ZJUT-EIFD were demonstrated. ZJUT-EIFD can serve benchmarks and interoperability tests for EF-IF research, which will promote the research and development of EF and IF.

Traditional technique application v/s YOLOv5 model for marine underwater objects detection by computer vision method

The paper compares the effectiveness of two methods application for detecting and recognizing micro-objects in an aqua environment on the example of plankton using neural networks and various technologies and developed in different programming languages. At first the traditional detection method was investigated and applied based on the extraction of Gabor and multilayer perceptron features, realized in MATrixLABoratory (MATLAB) language. Secondly, YOLOv5 (an acronym for ‘You only look once’), as a single-stage neural network was used which was implemented in Python language. The results of the work of these methods for the plankton detection are presented. Accuracy and completeness metrics are calculated to determine best of two methods. Images with the recognition result, programmatically calculated performance metrics were obtained after using the detection methods. The study was conducted on the effectiveness of method applications for realtime recognition using short video images. In conclusion it is stated that the YOLOv5 model has shown significant superiority over the traditional method in the task of detecting marine objects, in particular plankton. Its accuracy was 30% higher; the completeness of object detection was 27% higher.

An Augmented Classroom Teaching System based on AR and Facial Recognition

The goal of this study is to investigate the perception of the application of facial recognition and augmented reality in the classroom. HoloLens 2, an AR device, and Microsoft Azure cognitive services were utilized to enhance teachers' classroom perception by displaying learning analytics information on the AR device in real time. A focus group interview was conducted to collect initial feedback from teachers and students regarding the benefits, challenges, and design concerns related to the proposed system. A refined interface was designed based on the feedback from the participants.

Research on Pedestrian Crossing Decision Models and Predictions Based on Machine Learning.

Systematically and comprehensively enhancing road traffic safety using artificial intelligence (AI) is of paramount importance, and it is gradually becoming a crucial framework in smart cities. Within this context of heightened attention, we propose to utilize machine learning (ML) to optimize and ameliorate pedestrian crossing predictions in intelligent transportation systems, where the crossing process is vital to pedestrian crossing behavior. Compared with traditional analytical models, the application of OpenCV image recognition and machine learning methods can analyze the mechanisms of pedestrian crossing behaviors with greater accuracy, thereby more precisely judging and simulating pedestrian violations in crossing. Authentic pedestrian crossing behavior data were extracted from signalized intersection scenarios in Chinese cities, and several machine learning models, including decision trees, multilayer perceptrons, Bayesian algorithms, and support vector machines, were trained and tested. In comparing the various models, the results indicate that the support vector machine (SVM) model exhibited optimal accuracy in predicting pedestrian crossing probabilities and speeds, and it can be applied in pedestrian crossing prediction and traffic simulation systems in intelligent transportation.

De novo fabrication of higher arene ring incorporated contorted calix[2]phyrin(2.2.1.1.1) and its F- bound complex.

The rational design and synthesis of a novel contorted calix[2]phyrin(2.2.1.1.1) structure has been achieved, utilizing a terphenyl unit as a key building component. This terphenyl unit serves as a segment of the armchair periphery in a π-extended two-dimensional architecture. The resulting molecule exhibits remarkable properties, including the ability to self-assemble into solid-state supramolecular nanotubes. Additionally, it has demonstrated an affinity for complexation with fluoride anions, highlighting its potential for applications in molecular recognition and sensor technology. The incorporation of the terphenyl unit not only enhances the structural rigidity but also contributes to the unique electronic characteristics of the calix[2]phyrin.

Application of object detection and action recognition toward automated recognition of chemical experiments

Deep learning-based computer vision techniques enable object detection of chemical apparatus in images and action recognition of chemical experiment manipulations in videos.

Research on workflow recognition for liver rupture repair surgery.

Liver rupture repair surgery serves as one tool to treat liver rupture, especially beneficial for cases of mild liver rupture hemorrhage. Liver rupture can catalyze critical conditions such as hemorrhage and shock. Surgical workflow recognition in liver rupture repair surgery videos presents a significant task aimed at reducing surgical mistakes and enhancing the quality of surgeries conducted by surgeons. A liver rupture repair simulation surgical dataset is proposed in this paper which consists of 45 videos collaboratively completed by nine surgeons. Furthermore, an end-to-end SA-RLNet, a self attention-based recurrent convolutional neural network, is introduced in this paper. The self-attention mechanism is used to automatically identify the importance of input features in various instances and associate the relationships between input features. The accuracy of the surgical phase classification of the SA-RLNet approach is 90.6%. The present study demonstrates that the SA-RLNet approach shows strong generalization capabilities on the dataset. SA-RLNet has proved to be advantageous in capturing subtle variations between surgical phases. The application of surgical workflow recognition has promising feasibility in liver rupture repair surgery.

Automatic Generation of Assembly Instructions by Analyzing Process Recordings – A Concept Overview

The manual generation of instructions for complex assembly processes requires time and effort. Therefore, many companies refrain from documenting assembly processes and rely on the tacit expert knowledge of their employees. Due to demographic changes, the capturing of this implicit knowledge becomes increasingly relevant to prevent expertise loss. In this paper, we present a data acquisition and processing concept for the automated generation of assembly instructions. First, the assembly process is recorded using cameras. Then, the recordings are analysed combining multiple machine vision applications for object and activity recognition. Finally, the results are merged and displayed as digital assembly instructions.

Computationally intelligent real-time security surveillance system in the education sector using deep learning.

Real-time security surveillance and identity matching using face detection and recognition are central research areas within computer vision. The classical facial detection techniques include Haar-like, MTCNN, AdaBoost, and others. These techniques employ template matching and geometric facial features for detecting faces, striving for a balance between detection time and accuracy. To address this issue, the current research presents an enhanced FaceNet network. The RetinaFace is employed to perform expeditious face detection and alignment. Subsequently, FaceNet, with an improved loss function is used to achieve face verification and recognition with high accuracy. The presented work involves a comparative evaluation of the proposed network framework against both traditional and deep learning techniques in terms of face detection and recognition performance. The experimental findings demonstrate that an enhanced FaceNet can successfully meet the real-time facial recognition requirements, and the accuracy of face recognition is 99.86% which fulfills the actual requirement. Consequently, the proposed solution holds significant potential for applications in face detection and recognition within the education sector for real-time security surveillance.

Efficient Multi-Domain Text Recognition Deep Neural Network Parameterization With Residual Adapters

Recent advancements in deep neural networks have markedly enhanced the performance of computer vision tasks, yet the specialized nature of these networks often necessitates extensive data and high computational power. Addressing these requirements, this study presents a novel neural network model adept at optical character recognition (OCR) across diverse domains, leveraging the strengths of multi-task learning to improve efficiency and generalization. The model is designed to achieve rapid adaptation to new domains, maintain a compact size conducive to reduced computational resource demand, ensure high accuracy, retain knowledge from previous learning experiences, and allow for domain-specific performance improvements without the need to retrain entirely. Rigorous evaluation on open datasets has validated the model’s ability to significantly lower the number of trainable parameters without sacrificing performance, indicating its potential as a scalable and adaptable solution in the field of computer vision, particularly for applications in optical text recognition

Construction of a pillar[5]arene-based supramolecular chiral polymer linked to aminophosphine salt for chiral recognition of enantiomers of mandelic acid.

In recent years, supramolecular chirality has been greatly developed in asymmetric synthesis, chiral sensing and other research fields, but its application in molecular chiral recognition has not been extensively studied. In this paper, L-Boc-tyrosine methoxyester and phosphorus chloride salts were introduced into the framework of pillar[n]arene, and a pillar[5]arene-based supramolecular chiral polymer L-TPP-P was constructed. The supramolecular polymer had stable supramolecular chiral properties and could be used as a chiral solvation reagent for chiral recognition of mandelic acid MA. The molar ratio method and Scatchard plot showed that the complexation ratio of L-TPP-P (pillar[5]arene monomer as the reference object) and MA was 1 : 1, and the complexation constants of L-TPP-P with R-MA and S-MA were 4.51 × 105 M-1 and 6.5 × 104 M-1, respectively. The significant affinity difference of L-TPP-P for different enantiomers of MA showed the excellent chiral recognition and stereoselectivity of pillar[5]arene-based supramolecular polymers for MA. This study provides a new idea for a novel supramolecular polymer chiral recognition reagent or chiral recognition method.

Retraction Notice: Enhancing the Governance Capabilities Through Smart Technology: Scenario Application of Image Recognition and Its Effects in Chinese Local Governance

Retraction Notice: Enhancing the Governance Capabilities Through Smart Technology: Scenario Application of Image Recognition and Its Effects in Chinese Local Governance

AI Empowers Data Mining Models for Financial Fraud Detection and Prevention Systems

With the rapid development of Internet finance, the problem of financial fraud has become increasingly prominent, which has brought severe challenges to the security and stability of the financial industry. This paper aims to use artificial intelligence (AI) technology to empower data mining models to build an efficient financial fraud detection and prevention system. First of all, through the analysis of the development trend of Internet finance and the current situation of financial fraud, the limitations of traditional prevention and control measures in the face of big data and complex fraud modes are revealed. Secondly, the paper introduces the application prospect of AI technology in the financial field, especially the advantages and potential in data mining and fraud detection. Then, the data mining techniques adopted are discussed in detail, including the application of machine learning-based model recognition and artificial intelligence algorithms, as well as data preprocessing and feature engineering in big data environment. Further, the paper describes the system architecture design and key technology implementation, including model training and optimization, real-time monitoring and early warning solutions. Finally, through the in-depth analysis and evaluation of the experimental results, the paper verifies the remarkable effect and feasibility of the system in financial fraud detection and prevention, and provides a new technical guarantee and risk management strategy for the financial industry. The experimental results show that the model has achieved good results in the actual data in the financial field, and provides effective technical guarantee and support for financial security.

A Study on the Innovation of Traditional Drama Performance Forms by Intelligent Media Based on AI-Assisted Analysis Framework

Abstract This paper investigates the impact of contemporary entertainment preferences among youth, who increasingly favor novel forms over traditional theater arts, driven by the rapid pace of modern life. Specifically, it explores how action recognition technology can transform theatrical performance forms through the lens of intelligent media. Utilizing the inertial sensor MPU9250, the study mathematically captures the action characteristics of theatrical performances using quaternion algebra. This process involves preprocessing steps that set the stage for coordinate system transformations. Furthermore, the research constructs a database of theatrical morphology by classifying and storing morphological features derived from a human 3D skeleton dataset. The effectiveness of motion recognition technology in enhancing theatrical performances is then empirically tested and analyzed. The results reveal that this technology successfully identifies morphological movements in traditional theater with accuracies of 74.42%, 73.77%, and 64.45%, respectively. An additional pedagogical assessment highlights the significant role of motion recognition technology in educational contexts. In particular, the study notes a substantial difference in the ‘imagination’ dimension scores between control and experimental groups. The control group scored an average of 6.67, whereas the experimental group achieved 8.07, indicating enhanced engagement and cognitive stimulation through the application of motion recognition. These findings underscore the potential of such technology to offer theoretical and practical insights for the revitalization and progressive development of traditional theatrical performance forms.

Deep Optimized Broad Learning System for Applications in Tabular Data Recognition.

The broad learning system (BLS) is a versatile and effective tool for analyzing tabular data. However, the rapid expansion of big data has resulted in an overwhelming amount of tabular data, necessitating the development of specialized tools for effective management and analysis. This article introduces an optimized BLS (OBLS) specifically tailored for big data analysis. In addition, a deep-optimized BLS (DOBLS) network is developed further to enhance the performance and efficiency of the OBLS. The main contributions of this article are: 1) by retracing the network's error from the output space to the latent space, the OBLS adjusts parameters in the feature and enhancement node layers. This process aims to achieve more resilient representations, resulting in improved performance; 2) the DOBLS is a multilayered structure consisting of multiple OBLSs, wherein each OBLS connects to the input and output layers, enabling direct data propagation. This design helps reduce information loss between layers, ensuring an efficient flow of information throughout the network; and 3) the proposed methods demonstrate robustness across various applications, including multiview feature embedding, one-class classification (OCC), camera model identification, electroencephalogram (EEG) signal processing, and radar signal analysis. Experimental results validate the effectiveness of the proposed models. To ensure reproducibility, the source code is available at https://github.com/1027051515/OBLS_DOBLS.

Measurement of Forces Acting on Single T-Cell Receptors.

Molecular forces are increasingly recognized as an important parameter to understand cellular signaling processes. In the recent years, evidence accumulated that also T-cells exert tensile forces via their T-cell receptor during the antigen recognition process. To measure such intercellular pulling forces, one can make use of the elastic properties of spider silk peptides, which act similar to Hookean springs: increased strain corresponds to increased stress applied to the peptide. Combined with Förster resonance energy transfer (FRET) to read out the strain, such peptides represent powerful and versatile nanoscopic force sensing tools. In this paper, we provide a detailed protocol how to synthesize a molecular force sensor for application in T-cell antigen recognition and hands-on guidelines on experiments and analysis of obtained single molecule FRET data.