Rahaf Adam Alnuaimi, Ranem Khaled Almasalmeh, Sarah Adel Baker, Maryam Nasser Alsaiaari · 5 authors
In this paper, we aim to solve critical issues organizations face during attendance monitoring. Conventional log-in systems fail to effectively ensure successful attendance monitoring, and challenges such as user manipulation, social distancing making biometric devices obsolete, and other issues arise. To address these challenges, we propose a two-factor authentication system based on distance and face recognition. The system incorporates advanced geo-tracking tools and technologies with web3 features and double-factor authentication using face recognition technologies and accompanying distance monitoring devices and tools. Our system provides secure, adaptive, and advanced log-ins for employees and attendance monitoring for employers. The proposed system is scalable by simply accompanying more distance-tracking devices with no additional support systems required. It is a smart, user-friendly, and effective log-in system designed to optimize resource and time allocation for any organization. Compared to other two-factor authentication systems, our system is faster, more secure, and does not require central devices. It is also more friendly and flexible, offering a viable solution for maintaining a safe environment and easing procedures for employees and managers.
Anwar Ali Sathio, Shafiq Ahmed Awan, Ali Orangzeb Panhwar, Ali Aamir · 6 authors
During the COVID-19 pandemic, finding effective methods to prevent the spread of infectious diseases has become critical. One important measure for reducing the transmission of airborne viruses is wearing face masks but enforcing mask-wearing regulations can be difficult in many settings. Real-time and accurate monitoring of mask usage is needed to address this challenge. To do so, we propose a method for mask detection using a convolutional neural network (CNN) and blockchain technology. Our system involves training a CNN model on a dataset of images of people with and without masks and then deploying it on IoT-enabled devices for real-time monitoring. The use of blockchain technology ensures the security and privacy of the data and enables the efficient sharing of resources among network participants. Our proposed system achieved 99% accuracy through CNN training and was transformed into a blockchain-enabled network mechanism with QR validation of every node for authentication. This approach has the potential to be an effective tool for promoting compliance with mask-wearing regulations and reducing the risk of infection. We present a framework for implementing this technique and discuss its potential benefits and challenges
Blockchain is considered one of the most innovative techniques in the recent decade, this is because it takes advantage of the trustless, distributed, and tamper-proof. Researchers have established the existence of blockchain in regions from finance to health care records. They are also devoted to extending the application in other areas. This paper proposes an attendance system integrating blockchain and face recognition techniques. The attendance system consists of two parts, the face recognition component enables the system to automatically take attendance of a participant, which reduces complicated paperwork and saves time. Then, it stores attendance information in a blockchain-based database component at a specific time, providing a reliable and tamper-proof solution where no one is entitled to edit or delete data. We introduced an attendance system that utilizes a convolutional neural network for face recognition and stores the attendance data on the blockchain. The system shows decent resilience on tamper, and saves time for automatic attendance marks.
Face recognition and analysis
Blockchain Technology Applications and Security
Advanced Steganography and Watermarking Techniques
Non-fungible tokens (NFTs) are becoming very popular in a large number of applications ranging from copyright protection to monetization of both physical and digital assets. It is however a fact that NFTs suffer from a large number of security issues that create a lack of trust in solutions based on them. In this paper, we provide an overview of some of the most critical security challenges in media assets in form of visual content and then propose a specific solution for one among them, namely, secure person identification used in the context of KnowYour-Customer (KYC) with emphasis on liveness detection. The solution includes an authentication procedure that matches a selfie photo to a photograph of an identity document (ID). The system runs through a series of steps. First, detection is applied to extract faces from the selfie and the ID. Then a face comparison is performed to assess if they belong to the same person. While these two procedures are standard in KYC, a liveness check is also included so as to increase the security. The latter ensures that the user undergoing identity verification is in front of the camera and not a fraudster attempting to impersonate another individual. The system instructs the user to perform gestures such as waving hands or tilting head in front of the camera. The algorithmic detection of these actions during the live feed will reveal whether or not the user is carrying out the instructed activities. Performance of the proposed solution is then assessed under varying conditions.
Quang Nhat Tran, Benjamin Turnbull, Min Wang, Jiankun Hu
Biometric authentication is, over time, becoming an indispensable complementary component to traditional authentication methods that use passwords and tokens. As a result, the research interest in the protection techniques for the biometric template has also grown considerably. In this paper, we present a light-weight AI-based biometric authentication that operates based on the binary representation of a biometric instance. In details, a binary classifier will be trained using the binary strings that represent the intraclass and interclass biometric subjects. The Support Vector Machine and Multi-layer Perceptron Neural Network are chosen as the classifier to evaluate the fingerprint-based and iris-based authentication capability. Afterward, the authenticated biometric string is fed to a hash function to produce a hash value, which is to be used in a Zero-Knowledge-Proof Protocol for the purpose of privacy preservation. In order to improve the recognition of the classifier, we devise a simple yet efficient strategy to enhance the discriminativeness of the binary strings and name it the Composite Features Retrieval. We evaluated the proposed method with the four publicly available fingerprint datasets FVC2002-DB1, FVC2002-DB2, FVC2002-DB3, and FVC2004-DB2 and the iris dataset UBIRISv1. The promising performance shows this method's capability.
In the digital era where hacking and bypassing a system is easy, tampering of data is always possible leading to bad situations. Blockchain is used to store data which is near impossible to change or tamper with as it is very secure in nature. Voting as a process in any nation is an essential event and if votes get miscalculated by any external source it will be harmful. To avoid such kinds of situations and making it more comfortable blockchain technology comes in acknowledgment. This paper proposes a decentralized national e-voting system based on blockchain technology. It includes an admin panel to schedule the voting, manage candidates and declare the results. The web application will provide the users with an interface to enter their Aadhar card ID (text input) and a photo of themselves at the time of voting. The eligibility of the voter will be checked at the time they enter their Aadhar card ID. Eligible voter's phone numbers will be verified via One Time Password (OTP). After voter verification, individual voters will be considered eligible for voting. During voting, voters will be monitored through a webcam/front camera. The votes will be stored in a blockchain and any tampering would be detected easily. The address and the corresponding constituency will be checked in the backend. Voting results will be declared on a specified date and will be handled by the admin. The results will be displayed graphically with various options to choose from and will also include past results and statistics.
Face recognition and analysis
Advanced Steganography and Watermarking Techniques
Biometric verification has been widely deployed in current authentication solutions as it proves the physical presence of individuals. To protect the sensitive biometric data in such systems, several solutions have been developed that provide security against honest-but-curious (semi-honest) attackers. However, in practice attackers typically do not act honestly and multiple studies have shown drastic biometric information leakage in such honest-but-curious solutions when considering dishonest, malicious attackers. In this paper, we propose a provably secure biometric verification protocol to withstand malicious attackers and prevent biometric data from any sort of leakage. The proposed protocol is based on a homomorphically encrypted log likelihood-ratio-based (HELR) classifier that supports any biometric modality (e.g. face, fingerprint, dynamic signature, etc.) encoded as a fixed-length real-valued feature vector and performs an accurate and fast biometric recognition. Our protocol, that is secure against malicious adversaries, is designed from a protocol secure against semi-honest adversaries enhanced by zero-knowledge proofs. We evaluate both protocols for various security levels and record a sub-second speed (between $0.37$s and $0.88$s) for the protocol against semi-honest adversaries and between $0.95$s and $2.50$s for the protocol secure against malicious adversaries.
Biometric verification has been widely deployed in current authentication solutions as it proves the physical presence of individuals. Several solutions have been developed to protect the sensitive biometric data in such systems that provide security against honest-but-curious (a.k.a. semi-honest) attackers. However, in practice, attackers typically do not act honestly and multiple studies have shown severe biometric information leakage in such honest-but-curious solutions when considering dishonest, malicious attackers. In this paper, we propose a provably secure biometric verification protocol to withstand malicious attackers and prevent biometric data from any leakage. The proposed protocol is based on a homomorphically encrypted log likelihood-ratio (HELR) classifier that supports any biometric modality (e.g., face, fingerprint, dynamic signature, etc.) encoded as a fixed-length real-valued feature vector. The HELR classifier performs an accurate and fast biometric recognition. Furthermore, our protocol, which is secure against malicious adversaries, is designed from a protocol secure against semi-honest adversaries enhanced by zero-knowledge proofs. We evaluate both protocols for various security levels and record a sub-second speed (between 0.37s and 0.88s) for the protocol secure against semi-honest adversaries and between 0.95s and 2.50s for the protocol secure against malicious adversaries.
Shashank Agrawal, Saikrishna Badrinarayanan, Pratyay Mukherjee, Peter Rindal
We use biometrics like fingerprints and facial images to identify ourselves to our mobile devices and log on to applications everyday. Such authentication is internal-facing: we provide measurement on the same device where the template is stored. If our personal devices could participate in external-facing authentication too, where biometric measurement is captured by a nearby external sensor, then we could also enjoy a frictionless authentication experience in a variety of physical spaces like grocery stores, convention centers, ATMs, etc. The open setting of a physical space brings forth important privacy concerns though. We design a suite of secure protocols for external-facing authentication based on the cosine similarity metric which provide privacy for both user templates stored on their devices and the biometric measurement captured by external sensors in this open setting. The protocols provide different levels of security, ranging from passive security with some leakage to active security with no leakage at all. With the help of new packing techniques and zero-knowledge proofs for Paillier encryption -- and careful protocol design, our protocols achieve very practical performance numbers. For templates of length 256 with elements of size 16 bits each, our fastest protocol takes merely 0.024 seconds to compute a match, but even the slowest one takes no more than 0.12 seconds. The communication overhead of our protocols is very small too. The passive and actively secure protocols (with some leakage) need to exchange just 16.5KB and 27.8KB of data, respectively. The first message is designed to be reusable and, if sent in advance, would cut the overhead down to just 0.5KB and 0.8KB, respectively.
Inzamam Mashood Nasir, Muhammad Attique Khan, Ammar Armghan, Muhammad Younus Javed
Real-time applications like object detection, fire detection, face recognition and cancer detection are solely or partially relying on deep learning algorithms. Any tempering in these models can cause huge damages in many ways, therefore an utter need to secure these deep learning models is critically required. Blockchain technology has gained a wide popularity in tractability and security. In this article, the properties of blockchain are applied on the CNN models to produce secure CNN models. Each layer of a CNN model relates to a block, which contains the hash keys, public and private keys of their neighbors, while there exists a ledger block, which contains the detailed information about each layer of the model. The proposed SCNN model is tested using SVGG19 and SInceptionV3 models on publicly available datasets, which provides satisfactory results.
National security is a top priority to mitigate intrusions and criminal acts. Governments require robust national surveillance system that can cover all geographical areas, including the blind spots that may hold violence and criminal incidents' triggers i.e. malls, stadiums, airports, and other key sites. Integrating existing surveillance infrastructures rather than creating centralized solutions will have great potential on scalability as well as providing more liberal framework that is not run by a single point of control. However, this definitely requires establishing secure communication and mutual trust amongst these entities, which is a real challenge. Towards this end, we propose an efficient smart surveillance architecture that combines machine learning and Blockchain technologies to facilitate the exchange of relevant surveillance events as admitted transactions into a permissioned Hyperledger fabric Blockchain. We conducted comprehensive analysis to demonstrate the feasibility of blockchain and the efficiency of the machine learning-based face recognition and matching for real-time surveillance of suspects using heterogeneous surveillance infrastructure. The proposed architecture proved scalability and real-time behavior after putting the system through multiple test cases. With very high matching accuracy, and end-to-end latency of less than 12.8 seconds, the system proves to be scalable, and fast enough for a smart surveillance use case.
Kevin Putra Dirgantoro, Jae Min Lee, Dong‐Seong Kim
This paper proposes a face recognition for security system based on artificial intelligence and edge computing with a limited dataset. Generative adversarial networks (GANs) are used to manipulate the dataset in order to overcome the accuracy issue of the limited dataset. The average accuracy of GANs outperforms the limited dataset up to 92.79%. Edge computing is used to overcome a high latency of cloud computing with Jetson Nano board, which produces an average of 8.8 frame-per-second. Furthermore, the detected face will make a payment using a smart contract to the blockchain network with low static difficulty to open a gate or door's lock. In comparison, a low static difficulty outperforms the Proof-of- Work consensus algorithm in terms of transaction time around 33-39 milliseconds.
Face recognition and analysis
Biometric Identification and Security
Advanced Steganography and Watermarking Techniques
Muhammad Zahid Khan, Muhammad Zahid Khan, Muhammad Zahid Khan, Muhammad Usman Khan · 6 authors
Studies have been actively conducted on analyzing the driver's behavior inside the vehicle premises. Moreover, the transmission of the tempered proof multimedia content is also a major point of interest for the research community. At present, most of the techniques for detecting the distracted behavior of the driver is based on the detection of different face attributes like eyes and head posture etc, by using the traditional hand crafted features. In this paper we propose the deep learning based algorithm using the Convolution Neural Network. The proposed algorithm is independent of feature extraction of the specific parts, instead, it automatically picks the best features specific to the problem. We have utilized the State Form Distracted Driver Detection dataset to train our proposed algorithm. Furthermore, this paper also proposes a secure and tempered proof multimedia transaction. Original video data may be edited and fabricated with the false information. Multimedia blockchain can be helpful in tackling this problem. We have used Secure Hashing Algorithm (SHA‐256) algorithm for extracting the hashes of multimedia content. By utilizing the blockchain, we safely transmit the tempered proof video data coming from inside the vehicle, automatically detecting abnormal activities with our deep learning based algorithm. So, this paper combines the deep learning algorithms with blockchain techniques which is novel in research. Comparison between the results of proposed algorithm with the current state of the art work shows that proposed algorithm outperforms by achieving 86.02% accuracy on the test data.
Oscar Delgado-Mohatar, Julián Fiérrez, Rubén Tolosana, Rubén Vera-Rodríguez
We explore practical tradeoffs in blockchain-based biometric template storage. We first discuss opportunities and challenges in the integration of blockchain and biometrics, with emphasis in biometric template storage and protection, a key problem in biometrics still largely unsolved. Blockchain technologies provide excellent architectures and practical tools for securing and managing the sensitive and private data stored in biometric templates, but at a cost. We explore experimentally the key tradeoffs involved in that integration, namely: latency, processing time, economic cost, and biometric performance. We experimentally study those factors by implementing a smart contract on Ethereum for biometric template storage, whose cost-performance is evaluated by varying the complexity of state-of-the-art schemes for face and handwritten signature biometrics. We report our experiments using popular benchmarks in biometrics research, including deep learning approaches and databases captured in the wild. As a result, we experimentally show that straightforward schemes for data storage in blockchain (i.e., direct and hash-based) may be prohibitive for biometric template storage using state-of-the-art biometric methods. A good cost-performance tradeoff is shown by using a blockchain approach based on Merkle trees.
Blockchain technology has attracted a lot of attention in the previous years as a secure way to protect transactions in different processes. It has been particularly used to define cryptocurrencies. While inherently secure against classical single node attacks, the blockchain cryptocurrencies have recently been subject to attacks by malwares able to capture a single user wallet and its included keys. In this work we propose the use of biometric cryptosystems to control the access to the wallets on single machines. After a brief description of the blockchain, the cryptocurrencies and the possible attacks, the paper describes the use of convolutional neural network face recognition as a tool to extract biometric features that help in a key binding approach to protect the personal data in the wallet. Experiments have been conducted on three independent face datasets and the results obtained are satisfactory. The equal error rate between false acceptance and false rejection is negligible when testing on images from the same dataset used for the training of the convolutional neural network. This generalizes well when experimenting on two other independent datasets. These results prove that face cryptosystems can be used to protect the access on sensitive data existing in the wallets of many cryptocurrencies.