Utilizing deep neural nets for an embedded ECG-based biometric authentication system

Page, Adam; Kulkarni, Amey; Mohsenin, Tinoosh

doi:10.1109/biocas.2015.7348372

Cited by 69 publications

(28 citation statements)

References 12 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…DNNs have also been applied for the recognition of other biometric characteristics, such as signature, 59 finger vein, 60,61 or electrocardiography. 62 However, to the best of our knowledge there is no complete study of the possibilities offered by deep learning when applied to the fingerprint classification problem. This paper aims to provide a first systematic study on the field, to analyze strengths and weaknesses of DNNs in this context.…”

Section: Fingerprint Classification With Deep Neural Networkmentioning

confidence: 99%

On the use of convolutional neural networks for robust classification of multiple fingerprint captures

Peralta

Triguero

García

et al. 2017

Int. J. Intell. Syst.

View full text Add to dashboard Cite

Fingerprint classification is one of the most common approaches to accelerate the identification in large databases of fingerprints. Fingerprints are grouped into disjoint classes, so that an input fingerprint is compared only with those belonging to the predicted class, reducing the penetration rate of the search. The classification procedure usually starts by the extraction of features from the fingerprint image, frequently based on visual characteristics. In this work, we propose an approach to fingerprint classification using convolutional neural networks, which avoid the necessity of an explicit feature extraction process by incorporating the image processing within the training of the classifier. Furthermore, such an approach is able to predict a class even for low-quality fingerprints that are rejected by commonly used algorithms, such as FingerCode. The study gives special importance to the robustness of the classification for different impressions of the same fingerprint, aiming to minimize the penetration in the database. In our experiments, convolutional neural networks yielded better accuracy and penetration rate than state-of-the-art classifiers based on explicit feature extraction. The tested networks also improved on the runtime, as a result of the joint optimization of both feature extraction and classification. wileyonlinelibrary.com/journal/int

show abstract

Section: Fingerprint Classification With Deep Neural Networkmentioning

confidence: 99%

On the use of convolutional neural networks for robust classification of multiple fingerprint captures

Peralta

Triguero

García

et al. 2017

Int. J. Intell. Syst.

View full text Add to dashboard Cite

show abstract

“…In some of the reported works, the proportion of training data against testing data is benchmarked from 70% to 90% [23,24]. While having the 70% of the data may seem appropriate, it is necessary to explore the strength of Deep Learning with the aim to reduce the training data required.…”

Section: Classification Performance On Proportion Of Trainingmentioning

confidence: 99%

“…Apart from statistical methodologies, there are works that use Neural Network approaches. The work in [23] uses neural network to identify and extract QRS complex which is use as a classifier in Deep Neural Net. While the result achieved an impressive 99.54% accuracy with a database of 90 individuals and used 70% of the data for training, it is to note that signals used in the studies were acquired under resting states.…”

Section: Introductionmentioning

confidence: 99%

Deep Multiview Heartwave Authentication

Lim

Woo

Dlay

et al. 2019

IEEE Trans. Ind. Inf.

View full text Add to dashboard Cite

This paper presents a heartwave based authentication method that utilizes an ensemble of Deep Belief Networks (DBNs) under different parameters to increase the reliability of feature extraction. The multi-view outputs are further embed into a single view before inputting into a stacked DBN for classification. The result of the proposed novel architecture achieved a classification rate of 98.3% with 30% training data. Importantly, it is able to perform user classification using heartwave signals acquired under intense physical exercise where heartrate ranges from 50bpm to as high as 180bpm. Under extreme physical duress, the heartwave from individual experiences extreme morphological variations which render conventional classification approaches non-applicable.

show abstract

“…STT provides i) approximately 4X higher integration density than conventional Static Random Access Memory (SRAM) [11,17,15,21], ii) high retention times (even more than 10 years [9]), iii) high endurance (1016 writes, or 10 years of operation as L1 cache) [1], iv) near-zero leakage [18] with close to SRAM read performance, v) excellent thermal robustness 300oC, vi) soft error resilience, and vii) above all, STT cells are easy to integrate with the conventional CMOS fabrication process. To date, STT technology has been only used to break through the memory wall by implementing low-power, high-density on-chip memories.…”

Section: A Overview Of Stt Technologymentioning

confidence: 99%

Preventing design reverse engineering with reconfigurable spin transfer torque LUT gates

Winograd

Salmani

Mahmoodi

et al. 2016

2016 17th International Symposium on Quality Electronic Design (ISQED)

View full text Add to dashboard Cite

This paper presents a rigorous step towards design-for-assurance by employing the non-volatile spin transfer torque magnetic technology to design reconfigurable Look-Up-Tables logic components (NV-STT-based LUTs). Further, we introduce a novel security-driven STT-CMOS hybrid design flow that ensures the functionality of NV-STT-based LUTs cannot be determined in any manageable time, rendering any design reverse engineering attack ineffective. In addition, the flow ensures that there is minimum or no impact on design parametric constraints including performance, power and area.

show abstract

Utilizing deep neural nets for an embedded ECG-based biometric authentication system

Cited by 69 publications

References 12 publications

On the use of convolutional neural networks for robust classification of multiple fingerprint captures

On the use of convolutional neural networks for robust classification of multiple fingerprint captures

Deep Multiview Heartwave Authentication

Preventing design reverse engineering with reconfigurable spin transfer torque LUT gates

Contact Info

Product

Resources

About