Biometric Based Security Systems Research Articles

A Patch-Based CNN Built on the VGG-16 Architecture for Real-Time Facial Liveness Detection

Facial recognition is a prevalent method for biometric authentication that is utilized in a variety of software applications. This technique is susceptible to spoofing attacks, in which an imposter gains access to a system by presenting the image of a legitimate user to the sensor, hence increasing the risks to social security. Consequently, facial liveness detection has become an essential step in the authentication process prior to granting access to users. In this study, we developed a patch-based convolutional neural network (CNN) with a deep component for facial liveness detection for security enhancement, which was based on the VGG-16 architecture. The approach was tested using two datasets: REPLAY-ATTACK and CASIA-FASD. According to the results, our approach produced the best results for the CASIA-FASD dataset, with reduced HTER and EER scores of 0.71% and 0.67%, respectively. The proposed approach also produced consistent results for the REPLAY-ATTACK dataset while maintaining balanced and low HTER and EER values of 1.52% and 0.30%, respectively. By adopting the suggested enhanced liveness detection, architecture that is based on artificial intelligence could make current biometric-based security systems more secure and sustainable while also reducing the risks to social security.

Image fusion by discrete wavelet transform for multimodal biometric recognition

In today’s world, security plays a crucial role in almost all applications. Providing security to a huge population is a more challenging task. Biometric security is the key player in such type of situation. Using a biometric-based security system more secure application can be built because it is tough to steal or forge. The unimodal biometric system uses only one biometric modality where some of the limitations will arise. For example, if we use fingerprints due to oiliness or scratches, the finger recognition rate may reduce. In order to overcome the drawbacks of unimodal biometrics, multimodal biometric systems were introduced. In this paper, new multimodal fusion methods are proposed, where instead of merging features, database images are fused using discrete wavelet transform (DWT) technique. Face and signature images are fused, features are extracted from the fused image, an ensemble classifier is used for classification, and also experiments are conducted for finger vein and signature images.

A deep learning-based multimodal biometric system using score fusion

Recent trends in artificial intelligence tools-based biometrics have overwhelming attention to security matters. The hybrid approaches are motivated by the fact that they combine mutual strengths and they overcome their limitations. Such approaches are being applied to the fields of biomedical engineering. A biometric system uses behavioural or physiological characteristics to identify an individual. The fusion of two or more of these biometric unique characteristics contributes to improving the security and overcomes the drawbacks of unimodal biometric-based security systems. This work proposes efficent multimodal biometric systems based on matching score concatenation fusion of face, left and right palm prints. Multimodal biometric identification systems using convolutional neural networks (CNN) and k-nearest neighbors (KNN) are proposed and trained to recognize and identify individuals using multi-modal biometrics scores. Some popular biometrics benchmarks such as FEI face dataset and IITD palm print database are used as raw data to train the biometric systems to design a strong and secure verification/identification system. Experiments are performed on noisy datasets to evaluate the performance of the proposed model in extreme scenarios. Computer simulation results show that the CNN and KNN multi-modal biometric system outperforms most of the most popular up to date biometric verification techniques.

Biometric-Based Security System for Smart Riding Clubs

Horses, workers or riders need safety in a farm or a riding club. On account of the great value of the horse, the breeder needs to protect it from theft and disease. In this context, it is important to detect and recognize the identity of each worker or rider and horse for security reasons. In fact, this paper proposes a Smart Riding Club Biometric System (SRCBS) consisting in automatically detecting and recognizing horses as well as humans. The proposed system is based on the facial biometrics for a horse as well as the human gait biometrics due to their simplicity and intuitiveness in an uncontrolled environment. This work suggests a Siamese network based on DenseNet features for human gait recognition, a Sparse Neural Network (SNN) based on sparse features for horse face detection and a horse face recognition method based on Gabor features, LDA and SVM. Because of the unavailability of horse databases, this paper presents a new benchmark for horse detection and recognition in order to evaluate our proposed system. The proposed systems achieved an average accuracy of gait recognition equal to 95% in the 0° view, 100% in the 90° view and 98.90% in the 180° view on Casia-B dataset, an average precision equal to 90% for horse face detection and a recognition rate equal to 99.89% for horse face identification.

Region Centric Minutiae Propagation Measure Orient Forgery Detection with Finger Print Analysis in Health Care Systems

AbstractThe problem of forgery detection has been well studied and the forged finger prints produces highly impacting results in the biometric based security systems. There are many algorithms discussed earlier to detect forged finger prints. However, they suffer to achieve higher performance in terms of security. In this paper, a region centric minutiae propagation measure (RCMPM) based approach. First, the finger print image is read and removes the noisy points by applying the multi level Gabor filters. The Gabor filter has been applied in multiple level which helps to remove the noise from finger print image. The enhanced image is converted into number of integral image. The integral images are generated by splitting the image into number of tiny images according to the size of window. From the integral image produced, the island, dot, enclosure, bifurcation features are extracted. Extracted features are framed as feature vector and used to estimate the RCMPM measure. Based on the RCMPM measure, the presence of forged finger print has been identified and the same has been used to identify the region which has been modified. The accuracy of forged print detection has been improved and reduces the false classification ratio.

Tampering Reveal Technique for Iris Images

Nowadays, many biometric based security systems depend on the iris images for authentication because of their features and ease of use. However, storing and sharing these sensitive images through open access networks will expose them to tampering. In order to ensure the safety of the iris images, this paper presents a new tampering reveal technique based on watermarking in the transform domain. In the proposed technique, the iris region is selected and isolated using interactive segmentation process (ISA). The authentication bits have been generated from the iris region (IR) and embedded in the Slantlet transform coefficients of the remaining part of the iris image which has been named as non-iris region (NIR). The use of ISA ensures the intactness of IR because it has been excluded from the embedding process. Several experiments have been conducted to test the visual quality, capacity, payload, and the tampering reveal performance. The experiments proved the ability of the proposed technique to reveal and localize any tampering in IR, in addition, the difference between the original iris image and the watermarked iris image is imperceptible

Deep Convolutional Neural Network for Double-Identity Fingerprint Detection

Automatic human recognition using ubiquitous fingerprint sensors is the most widely used modality in modern biometric based security systems. The double-identity fingerprint is a fake fingerprint created by aligning two fingerprints for maximum ridge similarity and then joining them along an estimated cutline such that relevant features of both fingerprints are present on either sides of the cutline. The fake fingerprint containing the features of the criminal and his innocuous accomplice can be enrolled with an electronic machine readable travel document and later used to cross the automated border gates by claiming identity of the accomplice. In this letter, we have developed a deep convolutional neural network (CNN)-based patch-learning approach to estimate the cutline by training the network to identify and learn the pattern around the region of the joint fingerprint. This is a recent, new fingerprint alteration technique, and due to the unavailability of any such public database, we have generated a new database of 450 double-identity fingerprints. Experimental results show that the deep learning based approach is able to predict the cutline with an equal error rate, which is the best when compared with many other popular handcrafted features for double-identity fingerprint detection.

Low-cost fitness and activity trackers for biometric authentication

Abstract Fitness and activity tracking devices acquire, process and store rich behavioural data that are consumed by the end-user to learn health insights. This rich data source also enables a secondary use of being part of a biometric authentication system. However, there are many open research challenges with the use of data generated by fitness and activity trackers as a biometric source. In this article, the challenge of using data acquired from low-cost devices is tackled. This includes investigating how to best partition the data to deduce repeatable behavioural traits, while maximizing the uniqueness between participant datasets. In this exploratory research, 3 months’ worth of data (heart rate, step count and sleep) for five participants is acquired and utilized in its raw form from low-cost devices. It is established that dividing the data into 14-h segments is deemed the most suitable based on measuring coefficients of variance. Several supervised machine learning algorithms are then applied where the performance is evaluated by six metrics to demonstrate the potential of employing this data source in biometric-based security systems.

An Enhanced IRIS Output Technique (EIOT) for Biometric Security

Human biometric features form the base for many security applications which identify humans uniquely. Human eyes and specifically the Iris based identifications are regarded as highly reliable systems. Iris based systems when combined with cryptography have been able to present higher biometric based security systems. This paper presents an Iris based human identification system called EIOT which can enhance biometric security. It is a set of unique sequential steps followed in Iris recognition and can be implemented in human authentications and identifications.

Multi-Biometric Watermarking Scheme Based on Interactive Segmentation Process

Iris-based security systems are highly recommended because of their security and ease of use. Different watermarking techniques have been presented to provide security while exchanging or storing iris images. The previous iris image watermarking techniques have successfully reached their aims, however, they suffer from some limitations such as the distortions that are presented in the iris region because of the watermark embedding process, the limited embedding capacity, and the lack of robustness against unintentional attacks. On the other hand, nowadays, the biometric-based security systems have directed their interest towards multi-biometric techniques in order to improve the performance of the individual’s recognition process. This paper presents a new multi-biometric watermarking (MBW) scheme in which the features of the fingerprint image and some personal information are embedded in the iris image. To avoid the abovementioned limitations, the proposed scheme presents an interactive segmentation algorithm (ISA) and a Slantlet transform based watermark embedding method. The proposed ISA prevents any distortion in the iris region which is a beneficial feature for the iris image recognition process. The experimental results proved the efficiency of the proposed ISA in comparison with the Hough transform based methods in terms of accuracy, embedding capacity, and execution time. The proposed MBW scheme performs better in comparison with the state-of-the-art methods in terms of the intactness of the iris region, the robustness against unintentional attacks, and the watermark embedding capacity.

A Fuzzy Approach to Recognize Face Using Contourlet Transform

Face recognition addresses identification, verification, and authentication in biometric-based security systems. This work enhances the contrast and edges in face images and recognizes the face using contourlet transform and fuzzy rules. Contourlet transformed image provides multiscale and directional information. The transformed image is divided into low-pass image (low-frequency image) and band-pass image (high-frequency image). The low-pass image is enhanced using fuzzy-based histogram specification since it deals with contrast. Band-pass image contains detailed information about the edges of the image and are enhanced using fuzzy rules and morphological gradient operators. The proposed system achieves the accuracy rate of 99.81% and 99.35% on Yale-B and JAFEE dataset, respectively, which is better than the existing curvelet and wavelet transform-based recognition. The incorporation of fuzzy rules enhances the mean intensity value of the edges to 34.19, which is better than Canny, Sobel, Prewitt, Robert and Laplacian edge detection techniques. Finally Discriminant Correlation Analysis (DCA) feature level fusion is applied to fuse enhanced edge intensities and histogram features for Support Vector Machine (SVM) classification.