Novel deep neural network based pattern field classification architectures.

Field classification is a new extension of traditional classification frameworks that attempts to utilize consistent information from a group of samples (termed fields). By forgoing the independent identically distributed (i.i.d.) assumption, field classification can achieve remarkably improved accuracy compared to traditional classification methods. Most studies of field classification have been conducted on traditional machine learning methods. In this paper, we propose integration with a Bayesian framework, for the first time, in order to extend field classification to deep learning and propose two novel deep neural network architectures: the Field Deep Perceptron (FDP) and the Field Deep Convolutional Neural Network (FDCNN). Specifically, we exploit a deep perceptron structure, typically a 6-layer structure, where the first 3 layers remove (learn) a 'style' from a group of samples to map them into a more discriminative space and the last 3 layers are trained to perform classification. For the FDCNN, we modify the AlexNet framework by adding style transformation layers within the hidden layers. We derive a novel learning scheme from a Bayesian framework and design a novel and efficient learning algorithm with guaranteed convergence for training the deep networks. The whole framework is interpreted with visualization features showing that the field deep neural network can better learn the style of a group of samples. Our developed models are also able to achieve transfer learning and learn transformations for newly introduced fields. We conduct extensive comparative experiments on benchmark data (including face, speech, and handwriting data) to validate our learning approach. Experimental results demonstrate that our proposed deep frameworks achieve significant improvements over other state-of-the-art algorithms, attaining new benchmark performance.

Presentation Attack Face Image Generation Based on a Deep Generative Adversarial Network.

Although face-based biometric recognition systems have been widely used in many applications, this type of recognition method is still vulnerable to presentation attacks, which use fake samples to deceive the recognition system. To overcome this problem, presentation attack detection (PAD) methods for face recognition systems (face-PAD), which aim to classify real and presentation attack face images before performing a recognition task, have been developed. However, the performance of PAD systems is limited and biased due to the lack of presentation attack images for training PAD systems. In this paper, we propose a method for artificially generating presentation attack face images by learning the characteristics of real and presentation attack images using a few captured images. As a result, our proposed method helps save time in collecting presentation attack samples for training PAD systems and possibly enhance the performance of PAD systems. Our study is the first attempt to generate PA face images for PAD system based on CycleGAN network, a deep-learning-based framework for image generation. In addition, we propose a new measurement method to evaluate the quality of generated PA images based on a face-PAD system. Through experiments with two public datasets (CASIA and Replay-mobile), we show that the generated face images can capture the characteristics of presentation attack images, making them usable as captured presentation attack samples for PAD system training.

Early Expression Detection via Online Multi-Instance Learning With Nonlinear Extension.

Video-based facial expression recognition has received substantial attention over the past decade, while early expression detection (EED) is still a relatively new and challenging problem. The goal of EED is to identify an expression as quickly as possible after the expression starts and before it ends. This timely ability has many potential applications, ranging from human-computer interaction to security. The max-margin early event detector (MMED) is a well-known ranking model for early event detection. It can achieve competitive EED performance but suffers from several critical limitations: 1) MMED lacks flexibility in extracting useful information for segment comparison, which leads to poor performance in exploring the ranking relation between segment pairs; 2) the training process is slow due to the large number of constraints, and the memory requirement is also usually hard to satisfy; and 3) MMED is linear in nature, and hence may not be appropriate for data in a nonlinear feature space. To overcome these limitations, we propose an online multi-instance learning (MIL) framework for EED. In particular, the MIL technique is first introduced to generalize MMED, resulting in the proposed MIL-based EED (MIED), which is more general and flexible than MMED, since various instance construction and combination strategies can be adopted. To accelerate the training process, we reformulate MIED in the online setting and develop online multi-instance learning framework for EED (OMIED). To further exploit the nonlinear structure of the data distribution, we incorporate the kernel methods in OMIED, which results in the proposed online kernel multi-instance learning for early expression detection. Experiments on two popular and one challenging video-based expression data sets demonstrate both the efficiency and effectiveness of the proposed methods.

Mobile Phone Incentives for Childhood Immunizations in Rural India.

OBJECTIVES: Young children in resource-poor settings remain inadequately immunized. We evaluated the role of compliance-linked incentives versus mobile phone messaging to improve childhood immunizations. METHODS: Children aged ≤24 months from a rural community in India were randomly assigned to either a control group or 1 of 2 study groups. A cloud-based, biometric-linked software platform was used for positive identification, record keeping for all groups, and delivery of automated mobile phone reminders with or without compliance-linked incentives (Indian rupee Rs30 or US dollar $0.50 of phone talk time) for the study groups. Immunization coverage was analyzed by using multivariable Poisson regression. RESULTS: Between July 11, 2016, and July 20, 2017, 608 children were randomly assigned to the study groups. Five hundred and forty-nine (90.3%) children fulfilled eligibility criteria, with a median age of 5 months; 51.4% were girls, 83.6% of their mothers had no schooling, and they were in the study for a median duration of 292 days. Median immunization coverage at enrollment was 33% in all groups and increased to 41.7% (interquartile range [IQR]: 23.1%-69.2%), 40.1% (IQR: 30.8%-69.2%), and 50.0% (IQR: 30.8%-76.9%) by the end of the study in the control group, the group with mobile phone reminders, and the compliance-linked incentives group, respectively. The administration of compliance-linked incentives was independently associated with improvement in immunization coverage and a modest increase in timeliness of immunizations. CONCLUSIONS: Compliance-linked incentives are an important intervention for improving the coverage and timeliness of immunizations in young children in resource-poor settings.

Face Alignment With Deep Regression.

In this paper, we present a deep regression approach for face alignment. The deep regressor is a neural network that consists of a global layer and multistage local layers. The global layer estimates the initial face shape from the whole image, while the following local layers iteratively update the shape with local image observations. Combining standard derivations and numerical approximations, we make all layers able to backpropagate error differentials, so that we can apply the standard backpropagation to jointly learn the parameters from all layers. We show that the resulting deep regressor gradually and evenly approaches the true facial landmarks stage by stage, avoiding the tendency that often occurs in the cascaded regression methods and deteriorates the overall performance: yielding early stage regressors with high alignment accuracy gains but later stage regressors with low alignment accuracy gains. Experimental results on standard benchmarks demonstrate that our approach brings significant improvements over previous cascaded regression algorithms.

The biometric imaginary: (Dis)trust in a policy vacuum.

The decision in Europe to implement biometric passports, visas and residence permits was made at the highest levels without much consultation, checks and balances. Council regulation came into force relatively unnoticed in January 2005, as part of wider securitization policies urging systems interoperability and data sharing across borders. This article examines the biometric imaginary that characterizes this European Union decision, dictated by executive powers in the policy vacuum after 9/11 - a depiction of mobility governance, technological necessity and whom/what to trust or distrust, calling upon phantom publics to justify decisions rather than test their grounding. We consult an online blog we operated in 2010 to unravel this imaginary years on. Drawing on Dewey's problem of the public, we discuss this temporary opening of a public space in which the imaginary could be reframed and contested, and how such activities may shape, if at all, relations between politics, publics, policy intervention and societal development.

Robust and efficient biometrics based password authentication scheme for telecare medicine information systems using extended chaotic maps.

The Telecare Medicine Information Systems (TMISs) provide an efficient communicating platform supporting the patients access health-care delivery services via internet or mobile networks. Authentication becomes an essential need when a remote patient logins into the telecare server. Recently, many extended chaotic maps based authentication schemes using smart cards for TMISs have been proposed. Li et al. proposed a secure smart cards based authentication scheme for TMISs using extended chaotic maps based on Lee's and Jiang et al.'s scheme. In this study, we show that Li et al.'s scheme has still some weaknesses such as violation the session key security, vulnerability to user impersonation attack and lack of local verification. To conquer these flaws, we propose a chaotic maps and smart cards based password authentication scheme by applying biometrics technique and hash function operations. Through the informal and formal security analyses, we demonstrate that our scheme is resilient possible known attacks including the attacks found in Li et al.'s scheme. As compared with the previous authentication schemes, the proposed scheme is more secure and efficient and hence more practical for telemedical environments.

An efficient biometric and password-based remote user authentication using smart card for Telecare Medical Information Systems in multi-server environment.

The medical organizations have introduced Telecare Medical Information System (TMIS) to provide a reliable facility by which a patient who is unable to go to a doctor in critical or urgent period, can communicate to a doctor through a medical server via internet from home. An authentication mechanism is needed in TMIS to hide the secret information of both parties, namely a server and a patient. Recent research includes patient's biometric information as well as password to design a remote user authentication scheme that enhances the security level. In a single server environment, one server is responsible for providing services to all the authorized remote patients. However, the problem arises if a patient wishes to access several branch servers, he/she needs to register to the branch servers individually. In 2014, Chuang and Chen proposed an remote user authentication scheme for multi-server environment. In this paper, we have shown that in their scheme, an non-register adversary can successfully logged-in into the system as a valid patient. To resist the weaknesses, we have proposed an authentication scheme for TMIS in multi-server environment where the patients can register to a root telecare server called registration center (RC) in one time to get services from all the telecare branch servers through their registered smart card. Security analysis and comparison shows that our proposed scheme provides better security with low computational and communication cost.

Three-factor anonymous authentication and key agreement scheme for Telecare Medicine Information Systems.

Nowadays, with comprehensive employment of the internet, healthcare delivery services is provided remotely by telecare medicine information systems (TMISs). A secure mechanism for authentication and key agreement is one of the most important security requirements for TMISs. Recently, Tan proposed a user anonymity preserving three-factor authentication scheme for TMIS. The present paper shows that Tan's scheme is vulnerable to replay attacks and Denial-of-Service attacks. In order to overcome these security flaws, a new and efficient three-factor anonymous authentication and key agreement scheme for TMIS is proposed. Security and performance analysis shows superiority of the proposed scheme in comparison with previously proposed schemes that are related to security of TMISs.

The potential of using brain images for authentication.

Biometric recognition (also known as biometrics) refers to the automated recognition of individuals based on their biological or behavioral traits. Examples of biometric traits include fingerprint, palmprint, iris, and face. The brain is the most important and complex organ in the human body. Can it be used as a biometric trait? In this study, we analyze the uniqueness of the brain and try to use the brain for identity authentication. The proposed brain-based verification system operates in two stages: gray matter extraction and gray matter matching. A modified brain segmentation algorithm is implemented for extracting gray matter from an input brain image. Then, an alignment-based matching algorithm is developed for brain matching. Experimental results on two data sets show that the proposed brain recognition system meets the high accuracy requirement of identity authentication. Though currently the acquisition of the brain is still time consuming and expensive, brain images are highly unique and have the potential possibility for authentication in view of pattern recognition.

Recognizing disguised faces: human and machine evaluation.

Face verification, though an easy task for humans, is a long-standing open research area. This is largely due to the challenging covariates, such as disguise and aging, which make it very hard to accurately verify the identity of a person. This paper investigates human and machine performance for recognizing/verifying disguised faces. Performance is also evaluated under familiarity and match/mismatch with the ethnicity of observers. The findings of this study are used to develop an automated algorithm to verify the faces presented under disguise variations. We use automatically localized feature descriptors which can identify disguised face patches and account for this information to achieve improved matching accuracy. The performance of the proposed algorithm is evaluated on the IIIT-Delhi Disguise database that contains images pertaining to 75 subjects with different kinds of disguise variations. The experiments suggest that the proposed algorithm can outperform a popular commercial system and evaluates them against humans in matching disguised face images.

Trusted computing strengthens cloud authentication.

Cloud computing is a new generation of technology which is designed to provide the commercial necessities, solve the IT management issues, and run the appropriate applications. Another entry on the list of cloud functions which has been handled internally is Identity Access Management (IAM). Companies encounter IAM as security challenges while adopting more technologies became apparent. Trust Multi-tenancy and trusted computing based on a Trusted Platform Module (TPM) are great technologies for solving the trust and security concerns in the cloud identity environment. Single sign-on (SSO) and OpenID have been released to solve security and privacy problems for cloud identity. This paper proposes the use of trusted computing, Federated Identity Management, and OpenID Web SSO to solve identity theft in the cloud. Besides, this proposed model has been simulated in .Net environment. Security analyzing, simulation, and BLP confidential model are three ways to evaluate and analyze our proposed model.