A Low-Cost Embedded Facial Recognition System for Door Access Control using Deep Learning

This project presents a Home Automation Security System that integrates face detection and recognition with RFID technology using a Raspberry Pi. The system aims to enhance security and access control in homes by automating the process of user authentication through facial recognition and RFID tags. The system utilizes a Raspberry Pi as the central control unit, a camera module for face detection and recognition, and an RFID reader to scan RFID tags for additional access control. The system functions by scanning an RFID tag, and upon detection of a valid tag, it activates the camera to capture the face of the person requesting access. The captured image is then processed using OpenCV and face_recognition libraries in Python to match the face with a pre-stored database of authorized individuals. If both the RFID tag and face match the stored data, the system grants access by performing automation actions such as unlocking doors or turning on lights. Keywords- Raspberry Pi, Pi Camera, Face Detection, Face Recognition, RFID Technology, RFID Tag Verification.

Efforts to leverage the benefits of Deep Learning(DL) models for performing inference in resource constrained embedded devices is very popular nowadays. Researchers worldwide are trying to come up with software and hardware accelerators that make pre-trained DL models suitable for running the inference phase in these devices. Apart from software and hardware accelerators, DL model partitioning and offloading to Cloud or Edge network servers is becoming more and more practicable with increasing importance of Edge Computing. DL inference workflow partitioning and offloading can augment software / hardware acceleration for improved latency and energy efficiency in resource constrained embedded systems. Efficacy of a computation offloading to system is dependent on proper profiling of timing and energy required for processing DL algorithms. In this work we implement a DL inference offloading system using Raspberry Pi 3 based robot vehicle with a hardware accelerator from Intel (Neural Compute Stick). We report the workload partitioning approach, detailed experimental results and performance improvements achieved. We demonstrate that the current approach of DL execution profiling without considering dynamic system load of the edge device results in sub-optimal partitioning of DL algorithm and provide a solution approach to that.

The article shows the relevance of the widespread use of computer vision technology on the example of face recognition as part of the access control and management system. The main methods that are used in the implementation of classical control systems and access control are considered. The scheme for the implementation of the access control with face recognition technology is described. The use of this technology makes it possible to increase the level of information security of enterprises and, as a result, reduce the possible financial damage from the implementation of attacks on their assets from illegitimate penetration into the protected area through the access control system using the passes of legal users.

The increased complexity of modern access control (AC) systems stems partly from the need to support diverse and multiple administrative domains. Systems engineering is a key technology to manage this complexity since it is capable of assuring that an operational system will adhere to the initial conceptual design and defined requirements. Specifically, the verification stage of an AC system should be based on techniques that have a sound and mathematical underpinning. Working on this assumption, model checking techniques are applied for the verification of predefined system properties, and thus, conducting a security analysis of a system. In this paper, we propose the utilization of automated and error-free model checking techniques for the verification of security properties in multi-domain AC systems. Therefore, we propose a formal definition in temporal logic of four AC system properties regarding secure inter-operation with Role-Based Access Control (RBAC) policies in order to be verified by using model checking. For this purpose, we demonstrate the implementation of a tool chain for expressing RBAC security policies, reasoning on role hierarchies and properly feeding the model checking process. The proposed approach can be applied in any RBAC model to efficiently detect non-conformance between an AC system and its security specifications. As a proof of concept, we provide examples illustrating the verification of the defined secure inter-operation properties in multi-domain RBAC policies.

A face recognition system for automatic door access control has been developed in this work with a view to providing a relatively more robust and foolproof access control which can provide better security and reduce human errors inherent in other conventional methods. The system was designed with machine learning and artificial intelligence to capture faces, train faces with machine mode, and run trained faces to grant access to the user. The system uses the RaspberryPi module, camera module, servo motor and the GSM module which were all incorporated into the fabricated building to make up the prototype developed to provide access control by means of facial biometrics. In order to grant access to registered users, various photos of the users were taken in different positions and expressions with proper illumination. The user’s face is been captured by the camera module and saved in the database with the help of Raspberry Pi Module. Good lighting condition and other favorable conditions helps the camera module to recognize faces and sends signal to the Raspberry Pi which processes these images and opens the door with the help of the servo motor. The developed prototype was used to train fifty (50) users. It granted access to all fifty (50) users when there was proper illumination and pose but five (5) and nine (9) users respectively were denied access due to challenges of poor illumination and pose variation.

A dormitory is a place where students live and live for a long time, and the safety of the dormitory can directly affect the quality of students’ studies and lives. In light of this, the study constructs a dynamic security planning system for college dormitory areas, enhancing their security and intelligence through the application of intelligent technology and multi-directional management coordination. Taking the access control system and fire monitoring system in the dormitory security prevention system as an example, based on the neural network perspective, we design the face recognition method of HOG-YOLO and the fire detection algorithm of improved YOLOv5 to realize, train, and test the two models, and evaluate the face recognition performance and fire target detection effect of the models. The results show that the face recognition method established in this paper is highly accurate and rapid, with a recognition accuracy of 95.7%, and it can complete recognition and detection under different facial angles. The fire detection algorithm’s FPS has been improved by 14.75%. The overall accuracy has improved by more than 1%, which has a good effect on fire monitoring. Planning the security system for the dormitory areas in higher education is a significant area of research for ensuring campus security, and access control and fire monitoring systems have significant application potential.

Introduction: Object detection has been an essential task in computer vision for decades, and modern developments in computer vision and deep learning have greatly increased the accuracy of detecting systems. However, the high computational requirements of deep learningbased object detection algorithms limit their applicability to resource-constrained systems, such as embedded devices. Methods: With the advent of Tiny Machine Learning (TinyML) devices, such as Raspberry Pi, it has become possible to deploy object detection systems on small, low-power devices. Due to their accessibility and cost, Tiny-ML devices, such as Raspberry Pi, a single-board tiny-ML device that is extremely well-liked, have recently attracted a lot of attention. Results: In this study, we present an enhanced SSD-based object detection approach and deploy the model using a tinyML device, i.e., Raspberry Pi. Conclusion: The proposed object detection model is lightweight and built utilizing Mobilenet- V2 as the underlying foundation.

Access control is the process of protecting resources by allowing or rejecting access to particular resources. It is an integral part of physical security systems and protects resources from unauthorized use. Over the past decade, there has been a huge demand and need for more sophisticated and secure access technologies. Scientists are developing cutting edge technologies to fulfill the requirements for secure access control experiences. Access control systems, using the latest biometric technologies such as fingerprint, facial recognition, iris recognition, DNA, and signature recognition have offered a higher level of security than the conventional systems. Biometrics plays a vital role in the identification and authentication phase of access control. Every biometric system has weaknesses and benefits. The system success depends on selecting the proper type of biometric modality to use and understanding how it affects system performance, cost, and limitations. Fingerprint holds the first place according to its popularity but facial recognition and iris recognition continue to grow for its accuracy and are considered as the future of biometric recognition. Besides biometric technology, other future possibilities for secure access control are- mobile access control, cloud-based security, and multi-factor authentication. This paper presents access control components, functions, and types. The aim of the paper is to discuss and analyze the latest biometric access control to secure sensitive information. This paper also provides an overview of access control, discusses the latest biometric technology, and presents the future trend of access control.

This paper shows a way to control a robot using Raspberry Pi and application built in the Android Platform. The purpose of this paper is to control the robot as upward, backward, right or left side and get a live video stream on Android phone from USB Cam by using Raspberry Pi. The Android phone and Raspberry Pi board is connected through WI-FI. As designed, the Android application is the command control center for the robot. The signals which is generated from Android app are sent to Raspberry pi and Raspberry pi receives these signals according to a program written in the python programming language in Raspberry pi. So the commands are processed and the robot is directed with user directive.

Innovative soft computing-enabled cloud optimization for next-generation IoT in digital twins

Face alignment and face recognition are two fundamental problems in the facial analysis community. For face alignment, it forms the basis for the accurate face recognition, age estimation, and facial expression recognition. For face recognition, it has been widely applied in various practical scenarios such as access control system, massive surveillance, human computer interaction, etc.. There mainly exist two lines of works in these two fields, namely holistic face alignment and recognition, and partial face alignment and recognition. Numerous holistic face alignment and recognition works

Face recognition technology has become more and more important in the field of biometric identification because of its advantages. A face recognition system combining hardware and software based on S3C2440 is presented in this paper. We adopt S3C2440 embedded development board as the hardware platform. This paper creates an embedded Linux software platform and our system is divided into four sections by their function: graphical user interface, image capture, face detection and face recognition. The whole system is tested as an access control system in practical situation. Experimental results show that our system has a good performance in both accuracy and efficiency of face recognition in access control.

Studies on two-factor authentication based on RFID and face recognition have been carried out on a large scale. However, these studies didn't discuss the way to overcome the weaknesses of face recognition authentication in the access control systems. In this study, two authentication factors, RFID and face recognition, were implemented using the LBP (Local Binary Pattern) algorithm to overcome weaknesses of face recognition authentication in the access control system. Based on the results of performance testing, the access control system has 100% RFID authentication and 80% face recognition authentication. The average time for the RFID authentication process is 0.03 seconds, the face recognition process is 6.3885 seconds and the verification of the face recognition is 0.1970 seconds. The access control system can still work properly after three days without being switched off. The results of security testing showed that the capabilities spoofing detection has 100% overcome the photo attack.

In traditional framework, mandatory access control (MAC) system and malicious software are run in kernel mode. Malicious software can stop MAC systems to be started and make it do invalid. This problem cannot be solved under the traditional framework if the operating system (OS) is comprised since malwares are running in ring0 level. In this paper, we propose a novel way to use hypervisors to protect kernel integrity and the access control system in commodity operating systems. We separate the access control system into three parts: policy management (PM), security server (SS) and policy enforcement (PE). Policy management and the security server reside in the security domain to protect them against malware and the isolation feather of the hypervisor can protect them from attacks. We add an access vector cache (AVC) between SS and PE in the guest OS, in order to speed up communication between the guest OS and the security domain. The policy enforcement module is retained in the guest OS for performance. The security of AVC and PE can be ensured by using a memory protection mechanism. The goal of protecting the OS kernel is to ensure the security of the execution path. We implement the system by a modified Xen hypervisor. The result shows that we can secure the security of the access control system in the guest OS with no overhead compared with modules in the latter. Our system offers a centralized security policy for virtual domains in virtual machine environments.

Although access control based on human face recognition has become popular in consumer applications, it still has several implementation issues before it can realize a stand-alone access control system. Owing to a lack of computational resources, lightweight and computationally efficient face recognition algorithms are required. The conventional access control systems require significant active cooperation from the users despite its non-aggressive nature. The lighting/illumination change is one of the most difficult and challenging problems for human-face-recognition-based access control applications. This paper presents the design and implementation of a user-friendly, stand-alone access control system based on human face recognition at a distance. The local binary pattern (LBP)-AdaBoost framework was employed for face and eyes detection, which is fast and invariant to illumination changes. It can detect faces and eyes of varied sizes at a distance. For fast face recognition with a high accuracy, the Gabor-LBP histogram framework was modified by substituting the Gabor wavelet with Gaussian derivative filters, which reduced the facial feature size by 40% of the Gabor-LBP-based facial features, and was robust to significant illumination changes and complicated backgrounds. The experiments on benchmark datasets produced face recognition accuracies of 97.27% on an E-face dataset and 99.06% on an XM2VTS dataset, respectively. The system achieved a 91.5% true acceptance rate with a 0.28% false acceptance rate and averaged a 5.26 frames/sec processing speed on a newly collected face image and video dataset in an indoor office environment.