NATIONAL UNIVERSITY HOCHIMINH CITY UNIVERSITY OF INFORMATION TECHNOLOGY FACULTY OF INFORMATION SYSTEMS QUACH BAO HUNG - 18520809 PHAM DUY HUNG - 18520805 GRADUATION THESIS ARTIFICIAL INTELLIGENCE-BASED FACE RECOGNITION FOR CUSTOMER/ EMPLOYEE MANAGEMENT IN MULTIPLE INDUSTRIAL INDUSTRY TO IMPROVE PERFORMANCE AND SECURITY BACHELOR OF ENGINEERING IN INFORMATION SYSTEMS THESIS ADVISOR PhD. DO TRONG HOP HO CHI MINH CITY, 2023 ACKNOWLEDGEMENT We, the group of authors for the graduation thesis on "ARTIFICIAL INTELLIGENCE-BASED FACE RECOGNITION FOR CUSTOMER/ EMPLOYEE MANAGEMENT IN MULTIPLE INDUSTRIAL INDUSTRY TO IMPROVE PERFORMANCE AND SECURITY" would like to certify that all the content and findings in this thesis were independently generated and truthfully reported. In many different businesses, we have used facial recognition technology that is based on artificial intelligence to manage customers and employees. In addition to exploring the uses and promise of this technology in enhancing performance and security across sectors, we have carefully examined face recognition techniques, models, and solutions.
We have developed face recognition models and created customer/employee management systems based on this platform by utilizing our knowledge and expertise in the domains of artificial intelligence, programming, and image processing. We conducted experiments, tested the system's performance and accuracy, and then analyzed and assessed the outcomes. We guarantee that all facts, figures, and references have been properly sourced. Creating this graduation thesis did not violate any laws or intellectual property rights.
Despite our best efforts, there may still be certain restrictions and omissions in the thesis. As a result, we truly appreciate any advice from academics and subject- matter specialists and hope that it will be received with understanding. Finally, we would want to convey our sincere appreciation to our professors, family, friends, and everyone else who has helped us accomplish this thesis. Ho Chi Minh City, June 20, 2023 TABLE OF CONTENTS CHAPTER | PROBLEM STATEMENT.
Introduction to the †OpiC. The importance and significance ofthe topic. Research objectives and questions. Scope and limitations of the OpIC.
Data Collection Methods. Data Analysis Techniques. CHAPTER 2 LITERATURE REVIEW. Facial Recognition Technology.
Overview of Facial Recognition Techniques and Algorithms. Facial Feature Extraction and Representation Methods. Al-Based Facial Recognition Applications in Customer/ Employee Management 8. Customer Identification and Personalization in the Retail Industry 20 2.
Employee Attendance Tracking and Access Control in the Manufacturing Industry. Security and Fraud Detection in the Banking Industry. Existing Solutions and Case Studies. Existing Face Recognition Model for Multiple Industries.
Evaluation of Performance, Efficiency, and Security. Choice of Face Recognition Model for Research. Data Collection and PreparatiOn. Selection and Acquisition of Facial Recognition Datasets.
Preprocessing Techniques for Noise Reduction and Image EnhanceIm€ni(L. Algorithm Selection and Development. Fine-tuning and Model Development. System Design and Implemenfation.
Architecture and Components of AI-Based Facial Recognition System 40 3. Integration with Existing Customer/ Employee Management Systems 46 3. Facial Recognition Applications: An Extensive Analysis. System Architecture and ComponenIs.
Functionality of the Facial Recognition Sysfem. Input and Output Mechanism. ccc Stướt59 CHAPTER 4 PRESENTATION, ASSESSMENT, DISCUSSION OF RESULTS 64 4. Performance evaluation indicators and evaluation criteria.
Experimental setup and configuratiOn. Results of evaluation and analysis 4. Accuracy, precision, good memory, and Fl-Score of face recognition ImOeÌS. Performance and speed of face recognition system.
Summary of results achieved. New contributions and new proposals .ă-occrctererrerrrrerrertrrerrrrrrrerrrrrrree 75 CHAPTER 6 DEVELOPMENT DIRECTION. Recommendations for further research directions. Real-world application development in other industries.
84 LIST OF IMAGES Figure 2. VGGFace model Architecture. cece eeseseeeeestseeneeseseeeneee 26 Figure 2. ArcFace model Architecture.
FaceNet model Architecture. Siamese Neural Networks General Strategy. Best convolutional architecture selected for the verification task. Picture of American swimmer Aaron Peisol inside the Labeled Faces in the Wild đataSe.
St tren 35 Figure 3. The Accuracy of the Model based on batch Size. The Accuracy of the Model based on Epochs. The Accuracy of the Model based on the learning rate.
Web/System Architecture. Relational Database Schemas. Data Flow DiagTraim. S11 ket it 3 Figure 3.
Permissions and Access Control .----¿-+- css+s+szx+cs+ 54 Figure 3. Rescaling the image to match the model requirement. Getting The Image with Corresponding ID In Google Drive. The API code used for recognizing an employee’s face.
Notification that the Check Out Process Is Successful. - - «c6 E11 TT H1 1101011211 tr hư 61 Figure 4. Calculate the model Accuracy, Precision, Recall rate, F1-Score, 22:90:30. Model Accuracy, Precision, Recall rate, F1-Score, FAR, FRR result ¬—.
Comparing two images and the time is taken using our model. Sending an image to my web API for recognition and receiving the T€SuÏ{S 1M tM. CPU Utilization Percentage .--¿-¿- - se s+c++xzkseererrkrkeree 70 Figure 4. Memory Utilization Percentage .----:-5-5¿5ccsxccersrxeree 70 LIST OF ABBREVIATIONS AI Artificial intelligence CNN Convolutional Neural Networks SVM Support Vector Machines SIFT Scale-Invariant Feature Transform PCA Principal Component Analysis LDA Linear Discriminant Analysis PCA Principal Component Analysis ToF Time-of-Flight 3DMM 3D Morphable Models AAM Active Appearance Models EBGM Elastic Bunch Graph Matching ABSTRACT This thesis investigates the use of artificial intelligence-based face recognition systems for customer and staff management in a variety of sectors in order to improve efficiency and security.
The study looks into the possible advantages of facial recognition technology and its influence on various industries. The goal is to solve the limits of traditional methods to customer and staff management by proposing an intelligent solution based on face recognition algorithms. The methodology employed involves a comprehensive analysis of existing facial recognition techniques, including deep learning models and feature extraction methods. The study evaluates the accuracy, recognition speed, and reliability of facial recognition systems using performance evaluation metrics such as accuracy rate, recognition speed, false positives, and false negatives.
Additionally, k-fold cross- validation and confusion matrix analysis are employed to assess the overall performance and reliability of the system. The findings highlight the significant advantages of facial recognition technology in customer and employee management. Accuracy is a crucial factor in ensuring reliable identification and authentication. The proposed facial recognition system achieves high accuracy rates, enabling effective management of customers and employees in various industry domains.
Furthermore, the system demonstrates efficient recognition speed, ensuring real-time processing and quick identification. The reliability of the system is evaluated through the measurement of false positives and false negatives, emphasizing the system's capability to accurately identify legitimate users and reject unauthorized individuals. This thesis makes significant contributions to the discipline. For starters, it gives a detailed review of the effectiveness of facial recognition technology in customer and personnel management.
In order to achieve efficient and secure procedures, the study highlights the necessity of accuracy, recognition speed, and dependability. Second, it recommends using k-fold cross-validation and confusion matrix analysis as evaluation methodologies to ensure a thorough and unbiased assessment of the system's performance. Finally, the findings of the study lead to a better understanding of the prospective uses of face recognition technology across sectors, emphasizing the importance of responsible deployment and adherence to ethical principles. Based on the findings ofthe research, this thesis suggests future research paths in the field of face recognition technology for customer and staff management.
Among these suggestions are: Integration with Other Biometric Modalities: Look into integrating face recognition technology with other biometric modalities, such as fingerprint or iris recognition, to improve the overall security and reliability of identification operations. This multi-modal method can provide greater authentication and lower the danger of identity fraud. Tailored Customer Experiences: Investigate the possibilities of facial recognition technology in improving tailored customer experiences. By detecting facial expressions and emotions, the technology can enable targeted marketing campaigns, individualized product suggestions, and customized services, ultimately enhancing consumer happiness and loyalty.
Considerations for Privacy and Ethics: Evaluate the impact of face recognition technology on user privacy and address issues about data protection and consent. Create strong privacy rules and methods to guarantee the responsible and transparent use of face recognition technology while taking legal and ethical factors into account. User acceptability and Perception Studies: Conduct user studies and surveys to learn about consumer and staff acceptability, views, and concerns about face recognition technologies. Understanding user attitudes and resolving their issues can aid in increasing system adoption and credibility.
By following these research areas, the field of face recognition technology will be able to progress further, enabling unique solutions for efficient and secure customer and staff management across a wide range of businesses. These suggestions provide opportunities for additional investigation, so contributing to the continued development and appropriate deployment of face recognition technology in real- world applications. CHAPTER 1 PROBLEM STATEMENT 1. Introduction to the topic Artificial intelligence (AI) and computer vision technologies have rapidly advanced, opening the door for creative solutions across a range of sectors.
The field of face recognition is one that has made great development. As a subset of AI, face recognition entails the identification and confirmation of people based on their visual traits. Effective customer and staff management are essential for the success of businesses across all industries in the fast-paced business world of today. Traditional means of identification and verification, including ID cards and passwords, can be time-consuming and frequently subject to security breaches.
As a result, there is an increasing demand for systems that are more safe and more effective and that may raise both productivity and security. The purpose of this study is to investigate how artificial intelligence-powered face recognition technology may be used in various businesses to manage employees and customers. This technology may provide accurate and automatic identification, authentication, and tracking of persons by utilizing the capabilities of AI algorithms, resulting to increased efficiency and security precautions. In the context of customer and staff management, the theoretical underpinnings and practical application of facial recognition systems are investigated in this thesis.
To develop a thorough grasp of the topic, it will examine the available literature, research projects, and pertinent works. Additionally, it will concentrate on building and creating a facial recognition system that complies with the unique demands and difficulties experienced by various businesses. Deep investigation of the possible advantages, restrictions, and ethical issues related to facial recognition technology is one of the intended research outputs. It will also provide useful insights into the conception, creation, and use of such systems, as well as a comprehensive analysis of their functionality and efficiency.
This research intends to add to the body of knowledge by addressing the crucial components of face recognition for managing customers and employees as well as to offer helpful advice for businesses looking to improve their operational effectiveness and security measures. The literature review, methods, findings, discussion, and future directions of this research will all be covered in detail in the coming chapters of this thesis. Through this investigation, we seek to shed light on face recognition technology's disruptive potential and its CHAPTER in revolutionizing consumer and staff management across a variety of sectors. The importance and significance of the topic Researching and using artificial intelligence-based face recognition technology for customer and staff management has become extremely significant and influential at a time when technological developments and technical improvements are occurring quickly.
First of all, Al-based face recognition offers a rapid, precise, and automated way to identify and validate people. As a result, businesses may more quickly and easily confirm the identities of their clients and staff, increase productivity, and reduce identity-related mistakes. Secondly, using face recognition technology to manage customers and employees may make workplaces safer and more secure. The hazards associated with lost passwords, ID cards, or other authentication devices are considerably decreased by employing face characteristics for authentication.
This leads to enhanced personal information protection and access management to crucial systems, enhancing an organization's overall security and privacy. Thirdly, a variety of businesses may easily implement the use of face recognition technology for personnel and customer management. This technology may significantly enhance labor productivity, customer experience, and people management across a variety of industries, including banking, insurance, retail, entertainment, and manufacturing.