Sensing technologies place significant interest in the use of biometrics for the recognition and assessment of individuals. Pattern mining techniques have established a critical step in the progress of sensor-based biometric systems that are capable of perceiving, recognizing, and computing sensor data, being a technology that searches for the high-level information about pattern recognition from low-level sensor readings in order to construct an artificial substitute for human recognition. The design of a successful sensor-based biometric recognition system needs to pay attention to the different issues involved in processing variable data being—acquisition of biometric data from a sensor, data pre-processing, feature extraction, recognition, and/or classification, clustering, and validation. A significant number of approaches from image processing, pattern identification, and machine learning have been used to process sensor data. This paper aims to deliver the state-of-the-art summary and present strategies for utilizing the broadly utilized pattern mining methods in order to identify the challenges as well as future research directions of sensor-based biometric systems.

/pdf/pattern-mining-approaches-used-in-sensor-based-biometric-4u6nejbpd8.pdf

Pattern Mining Approaches Used in Sensor-Based Biometric Recognition: A Review

Internet-connected smart devices in, on, and around us, (e.g., embedded devices, wearable devices, and smart sensors) can collect human biometric features and facilitate identity authentication. Existing approaches are mainly based on pattern recognition and machine learning algorithms, which may not be capable of processing uncertain user information. Thus, focusing on the uncertainty of users’ identity, this article proposes a fuzzy authentication system based on neural network and extreme value analysis. Specifically, we utilize biometric gait information of human body recognition. Our proposed authentication system is designed to implicitly authenticate users based on their gait, and can detect uncertain users and reject the authentication of unknown users. The performance is evaluated using an open dataset of 153 volunteers, where we manage to achieve a recognition accuracy rate of 98.4% and an error rate of unauthorized users at 6%.

A Fuzzy Authentication System Based on Neural Network Learning and Extreme Value Statistics

Machine learning and Design of Experiments: Alternative approaches or complementary methodologies for quality improvement?

Wearable sensors-based gait recognition is an effective method to recognize people’s identity by recognizing the unique way they walk. Recently, the adoption of deep learning networks for gait recognition has achieved significant performance improvement and become a new promising trend. However, most of the existing studies mainly focused on improving the gait recognition accuracy while ignored model complexity, which make them unsuitable for wearable devices. In this study, we proposed a lightweight attention-based Convolutional Neural Networks (CNN) model for wearable gait recognition. Specifically, a four-layer lightweight CNN was first employed to extract gait features. Then, a novel attention module based on contextual encoding information and depthwise separable convolution was designed and integrated into the lightweight CNN to enhance the extracted gait features and simplify the complexity of the model. Finally, the Softmax classifier was used for classification to realize gait recognition. We conducted comprehensive experiments to evaluate the performance of the proposed model on whuGait and OU-ISIR datasets. The effect of the proposed attention mechanisms, different data segmentation methods, and different attention mechanisms on gait recognition performance were studied and analyzed. The comparison results with the existing similar researches in terms of recognition accuracy and number of model parameters shown that our proposed model not only achieved a higher recognition performance but also reduced the model complexity by 86.5% on average.

A Lightweight Attention-Based CNN Model for Efficient Gait Recognition with Wearable IMU Sensors.

As a novelist and the most secure biometric method, finger vein recognition has gained substantial significance and various pertinent researches have been reported in literature. However, it is difficult to extract a more reliable and accurate finger vein pattern due to the random noise, poor lighting, illumination variation, image deformation and blur. Furthermore, improper parameter settings of SVMs lead to poor classification accuracy and apparently, not much relevant research has been conducted on its optimal parameter setting. To alleviate these problems, this paper proposes an efficient finger vein recognition framework consisting of the hybrid Local Phase Quantization (LPQ) for robust feature extraction and Grey Wolf Optimization based SVM (GWO-SVM) to compute the best parameter combination of SVM for optimal results of binary classification. Finger vein features are first extracted by integrating LPQ, which is invariant to motion blur and deformation, with Local Directional Pattern (LDP), which is robust to random noise and illumination variation, to augment the recognition performance and reduce the computational time. Then, GWO-SVM is used for classification in order to maximize the classification accuracy by determining the optimal SVM parameters. The extensive experimental results indicate remarkable performance and significant enhancements in terms of recognition accuracy by the proposed framework compared to the existing techniques and prove the effectiveness of the proposed framework on four tested finger vein datasets. It has outperformed the typical SVM approach and kNCN-SRC two-stage methodology via achieving the recognition accuracy of 98% and equal error rate as low as 0.1020%.

Hybrid local phase quantization and grey wolf optimization based SVM for finger vein recognition

Biometrics, as a computer science field, can be understood as the discipline that study how to generate computer models of the physical (e.g. hand geometry, fingerprints, iris and so on) and behavioral (e.g. signature; a kind of behavior pattern) characteristics of the human being for single or several individuals identification. Usually, these characteristics are used to provide authentication information for security systems. However, some of these characteristics are hard to obtain in a properly way and it is necessary to use several algorithms both to process them and to use them on a security systems. In this sense, in this paper it is presented an overview of some Machine Learning approaches for biometric pattern recognition.

/pdf/survey-of-biometric-pattern-recognition-via-machine-learning-4elxh44s2s.pdf

Survey of biometric pattern recognition via machine learning techniques

This article describes the design of a smart bulb, conceived as one who is able to maintain, under certain variations of external lighting a desired level of illumination is presented. This level of lighting is adjustable from internet, using the concept of internet of things (IoT) to said bulb. The design consists of a programming card Raspberry pi 3, an LED bulb, and a power circuit under c programming language, a fuzzy controller is used to maintain the desired lighting level. In conclusion, it is possible to obtain a home automation system with control from internet which makes intelligent functional bulb-controlled lighting systems.

Mauricio Mauledeoux

Papers

Survey of biometric pattern recognition via machine learning techniques

Smart bulb for IoT