scholarly journals Fingerprint Classification through Standard and Weighted Extreme Learning Machines

2020 ◽  
Vol 10 (12) ◽  
pp. 4125
Author(s):  
David Zabala-Blanco ◽  
Marco Mora ◽  
Ricardo J. Barrientos ◽  
Ruber Hernández-García ◽  
José Naranjo-Torres
Keyword(s):  
High Performance ◽  
Regularization Parameter ◽  
Geometric Mean ◽  
Golden Ratio ◽  
Computational Cost ◽  
Extreme Learning Machines ◽  
Deep Convolutional Neural Networks ◽  
Identification Schemes ◽  
Learning Machines ◽  
Fingerprint Classification

Fingerprint classification is a stage of biometric identification systems that aims to group fingerprints and reduce search times and computational complexity in the databases of fingerprints. The most recent works on this problem propose methods based on deep convolutional neural networks (CNNs) by adopting fingerprint images as inputs. These networks have achieved high classification performances, but with a high computational cost in the network training process, even by using high-performance computing techniques. In this paper, we introduce a novel fingerprint classification approach based on feature extractor models, and basic and modified extreme learning machines (ELMs), being the first time that this approach is adopted. The weighted ELMs naturally address the problem of unbalanced data, such as fingerprint databases. Some of the best and most recent extractors (Capelli02, Hong08, and Liu10), which are based on the most relevant visual characteristics of the fingerprint image, are considered. Considering the unbalanced classes for fingerprint identification schemes, we optimize the ELMs (standard, original weighted, and decay weighted) in terms of the geometric mean by estimating their hyper-parameters (regularization parameter, number of hidden neurons, and decay parameter). At the same time, the classic accuracy and penetration-rate metrics are computed for comparison purposes with the superior CNN-based methods reported in the literature. The experimental results show that weighted ELM with the presence of the golden-ratio in the weighted matrix (W-ELM2) overall outperforms the rest of the ELMs. The combination of the Hong08 extractor and W-ELM2 competes with CNNs in terms of the fingerprint classification efficacy, but the ELMs-based methods have been demonstrated their extremely fast training speeds in any context.

Using Extreme Learning Machines and the Backprojection Algorithm as an Alternative to Reconstruct Electrical Impedance Tomography Images

2021 ◽  
pp. 16-27
Author(s):  
Juliana Carneiro Gomes ◽  
Maíra Araújo de Santana ◽  
Clarisse Lins de Lima ◽  
Ricardo Emmanuel de Souza ◽  
Wellington Pinheiro dos Santos
Keyword(s):  
Electrical Impedance Tomography ◽  
Electrical Impedance ◽  
Imaging Techniques ◽  
Low Cost ◽  
Computational Cost ◽  
Extreme Learning Machines ◽  
Impedance Tomography ◽  
Learning Machines ◽  
Electrical Potentials ◽  
Reconstruction Methods

Electrical Impedance Tomography (EIT) is an imaging technique based on the excitation of electrode pairs applied to the surface of the imaged region. The electrical potentials generated from alternating current excitation are measured and then applied to boundary-based reconstruction methods. When compared to other imaging techniques, EIT is considered a low-cost technique without ionizing radiation emission, safer for patients. However, the resolution is still low, depending on efficient reconstruction methods and low computational cost. EIT has the potential to be used as an alternative test for early detection of breast lesions in general. The most accurate reconstruction methods tend to be very costly as they use optimization methods as a support. Backprojection tends to be rapid but more inaccurate. In this work, the authors propose a hybrid method, based on extreme learning machines and backprojection for EIT reconstruction. The results were applied to numerical phantoms and were considered adequate, with potential to be improved using post processing techniques.

scholarly journals Online Learning of Oil Leak Anomalies in Wind Turbines with Block-Based Binary Reservoir

Electronics ◽  
2021 ◽  
Vol 10 (22) ◽  
pp. 2836
Author(s):  
Matteo Cardoni ◽  
Danilo Pietro Pau ◽  
Laura Falaschetti ◽  
Claudio Turchetti ◽  
Marco Lattuada
Keyword(s):  
High Speed ◽  
Computational Cost ◽  
Image Sensors ◽  
Extreme Learning Machines ◽  
Accurate Identification ◽  
Power Generator ◽  
Learning Machines ◽  
Anomaly Detector ◽  
Block Based ◽  
Realistic Images

The focus of this work is to design a deeply quantized anomaly detector of oil leaks that may happen at the junction between the wind turbine high-speed shaft and the external bracket of the power generator. We propose a block-based binary shallow echo state network (BBS-ESN) architecture belonging to the reservoir computing (RC) category and, as we believe, it also extends the extreme learning machines (ELM) domain. Furthermore, BBS-ESN performs binary block-based online training using fixed and minimal computational complexity to achieve low power consumption and deployability on an off-the-shelf micro-controller (MCU). This has been achieved through binarization of the images and 1-bit quantization of the network weights and activations. 3D rendering has been used to generate a novel publicly available dataset of photo-realistic images similar to those potentially acquired by image sensors on the field while monitoring the junction, without and with oil leaks. Extensive experimentation has been conducted using a STM32H743ZI2 MCU running at 480 MHz and the results achieved show an accurate identification of anomalies, with a reduced computational cost per image and memory occupancy. Based on the obtained results, we conclude that BBS-ESN is feasible on off-the-shelf 32 bit MCUs. Moreover, the solution is also scalable in the number of image cameras to be deployed and to achieve accurate and fast oil leak detections from different viewpoints.

scholarly journals Learning Anytime Predictions in Neural Networks via Adaptive Loss Balancing

2019 ◽  
Vol 33 ◽  
pp. 3812-3821 ◽  
Author(s):  
Hanzhang Hu ◽  
Debadeepta Dey ◽  
Martial Hebert ◽  
J. Andrew Bagnell
Keyword(s):  
Neural Networks ◽  
High Performance ◽  
Geometric Mean ◽  
Computational Cost ◽  
Data Sets ◽  
Adaptive Weights ◽  
Multiple Viewpoints ◽  
The Cost ◽  
Theoretical Considerations ◽  
Adaptive Weighted Sum

This work considers the trade-off between accuracy and testtime computational cost of deep neural networks (DNNs) via anytime predictions from auxiliary predictions. Specifically, we optimize auxiliary losses jointly in an adaptive weighted sum, where the weights are inversely proportional to average of each loss. Intuitively, this balances the losses to have the same scale. We demonstrate theoretical considerations that motivate this approach from multiple viewpoints, including connecting it to optimizing the geometric mean of the expectation of each loss, an objective that ignores the scale of losses. Experimentally, the adaptive weights induce more competitive anytime predictions on multiple recognition data-sets and models than non-adaptive approaches including weighing all losses equally. In particular, anytime neural networks (ANNs) can achieve the same accuracy faster using adaptive weights on a small network than using static constant weights on a large one. For problems with high performance saturation, we also show a sequence of exponentially deepening ANNs can achieve near-optimal anytime results at any budget, at the cost of a const fraction of extra computation.

scholarly journals High-Performance Extreme Learning Machines: A Complete Toolbox for Big Data Applications

IEEE Access ◽  
2015 ◽  
Vol 3 ◽  
pp. 1011-1025 ◽  
Author(s):  
Anton Akusok ◽  
Kaj-Mikael Bjork ◽  
Yoan Miche ◽  
Amaury Lendasse
Keyword(s):  
Big Data ◽  
High Performance ◽  
Extreme Learning Machines ◽  
Learning Machines ◽  
Big Data Applications

scholarly journals Matrix Information Geometry for Signal Detection via Hybrid MPI/OpenMP

Entropy ◽  
2019 ◽  
Vol 21 (12) ◽  
pp. 1184 ◽  
Author(s):  
Sheng Feng ◽  
Xiaoqiang Hua ◽  
Yongxian Wang ◽  
Qiang Lan ◽  
Xiaoqian Zhu
Keyword(s):  
Signal Detection ◽  
Message Passing ◽  
High Performance ◽  
Message Passing Interface ◽  
Dimensional Space ◽  
Geometric Mean ◽  
Computational Cost ◽  
Detection Performance ◽  
High Dimensional ◽  
The Matrix

The matrix information geometric signal detection (MIGSD) method has achieved satisfactory performance in many contexts of signal processing. However, this method involves many matrix exponential, logarithmic, and inverse operations, which result in high computational cost and limits in analyzing the detection performance in the case of a high-dimensional matrix. To address these problems, in this paper, a high-performance computing (HPC)-based MIGSD method is proposed, which is implemented using the hybrid message passing interface (MPI) and open multiple processing (OpenMP) techniques. Specifically, the clutter data are first modeled as a Hermitian positive-definite (HPD) matrix and mapped into a high-dimensional space, which constitutes a complex Riemannian manifold. Then, the task of computing the Riemannian distance on the manifold between the sample data and the geometric mean of these HPD matrices is assigned to each MPI process or OpenMP thread. Finally, via comparison with a threshold, the signal is identified and the detection probability is calculated. Using this approach, we analyzed the effect of the matrix dimension on the detection performance. The experimental results demonstrate the following: (1) parallel computing can effectively optimize the MIGSD method, which substantially improves the practicability of the algorithm; and (2) the method achieves superior detection performance under a higher dimensional HPD matrix.

Sign in / Sign up

Export Citation Format

Share Document