A novel approach aimed at elevating the performance of offline signature verification systems by harnessing the combined power of autoencoders and Convolutional Neural Networks (CNNs). Our evaluation encompassed Convo...
详细信息
ISBN:
(纸本)9798350360806;9798350360790
A novel approach aimed at elevating the performance of offline signature verification systems by harnessing the combined power of autoencoders and Convolutional Neural Networks (CNNs). Our evaluation encompassed Convolutional Neural Networks (CNN), K-Nearest Neighbours (KNN), Support Vector Machines (SVM), Gaussian Temporal Rule, Probabilistic Neural Networks, Multi-layered Perception, Long Short-Term Memory (LSTM), and various combinations thereof. Notably, our proposed autoencoder with Convolutional Neural Networks (AE with CNN) outshone all other approaches, achieving an impressive accuracy rate of 98.48%. While CNN displayed commendable performance at 89%, KNN and SVM fusion attained 78.50% accuracy, suggesting room for improvement in distinguishing genuine and forged signatures. The Gaussian Temporal Rule proved robust, with an accuracy of 91.20%, and Probabilistic Neural Networks and Multi-layered Perception with SVM reached accuracies of 92.06% and 91.67%, respectively. The introduction of LSTM in conjunction with SVM and KNN significantly enhanced accuracy to 95.40%, 95.20%, and 92.70%, respectively. Collectively, these findings provide valuable insights into the potential of AE with CNN as a leading solution for achieving highly accurate signature verification, particularly in contexts where the distinction between authentic and counterfeit signatures is critical.
This paper proposes a hardware accelerator for machine learning-based anomaly detection to enhance IoT security. Our model integrates Multilayer Perceptron (MLP) with Long Short-Term Memory (LSTM), utilizing an MLP-ba...
详细信息
ISBN:
(纸本)9798350330991;9798350331004
This paper proposes a hardware accelerator for machine learning-based anomaly detection to enhance IoT security. Our model integrates Multilayer Perceptron (MLP) with Long Short-Term Memory (LSTM), utilizing an MLP-based autoencoder and Isolation Forest algorithm for data dimensionality reduction and computational complexity reduction. Prototyped on a Zynq UltraScale+ XCZU9EG FPGA, our AE-LSTM model surpasses baseline MLP-only and LSTM-only models in resource utilization efficiency and detection accuracy. Compared to these baselines, it reduces parameters by 79.4% and 98% and LUT usage by 61.4% and 90.8%, respectively, while minimizing other resource utilization. Furthermore, power consumption is lowered to about 40% of the MLP-based model's consumption rate and 36% of the LSTM-based model's rate, with latency reduced to less than one-third from both baselines.
Time series anomaly detection is one of the critical tasks in time series analysis. There are many existing works using autoencoders for time series anomaly detection;however, autoencoders can not only reconstruct nor...
详细信息
ISBN:
(纸本)9798350349184;9798350349191
Time series anomaly detection is one of the critical tasks in time series analysis. There are many existing works using autoencoders for time series anomaly detection;however, autoencoders can not only reconstruct normal sequences but also reconstruct anomalous sequences sometimes, which leads to the poor effect of anomaly detection based on reconstruction error. At the same time, the real-world time series have the characteristic of temporal coupling, and the existing methods often cannot take into account the multiple modes of the time series, which leads to the low accuracy of time series anomaly detection. Aiming at the above problems, we propose a Time Series Anomaly Detection method incorporating Wavelet Decomposition and Temporal Decoupled autoencoder (WDAE). Specifically, WDAE first reshapes the original 1D time series into 2D data harboring different temporal patterns based on frequency components and independently extracts the different temporal features of the time series data. Then, the original data are reconstructed in the time domain by the autoencoder to obtain the reconstructed time-domain sequence, and in the frequency domain, the 2D data are subjected to the discrete wavelet transform and the inverse wavelet transform after weight adjustment to obtain the reconstructed frequency domain sequence, and the captured normal sequence pattern in the frequency domain restricts the latent feature space in the time domain so that the autoencoder can rebuild the normal sequence to avoid reconstructing the anomalous sequence well. Experiments on five publicly available data show that WDAE significantly improves the F1 value over related state-of-the-art baseline methods.
Analyzing colors in images is an approach to understanding semantic meaning. However, existing research often faces challenges due to limited dataset sizes or the need to create custom datasets. Insufficient data can ...
详细信息
ISBN:
(纸本)9798350385113;9798350385106
Analyzing colors in images is an approach to understanding semantic meaning. However, existing research often faces challenges due to limited dataset sizes or the need to create custom datasets. Insufficient data can lead to overfitting during model training and hinder generalization. To address this, we propose a two-stage machine learning method that leverages Kobayashi's Color Image Scale (CIS), a publicly available color image dataset, to enhance the predictive accuracy of color image classifier. In our method, we not only extract colors and image categories as features but also capture the xy-coordinates of color schemes from the CIS. These coordinates play a significant role in improving the accuracy of color image classification. Our two-stage learning method provides a straightforward and effective solution for enhancing predictive accuracy. Through our method, we achieve an impressive 97.36% accuracy in color image classification on the test dataset.
The polarimetric synthetic aperture radar (PolSAR) images contain fine characteristics and abstract spatial features, which attention to them can improve the classification accuracy. In this work, the residual convolu...
详细信息
ISBN:
(纸本)9798350350494;9798350350500
The polarimetric synthetic aperture radar (PolSAR) images contain fine characteristics and abstract spatial features, which attention to them can improve the classification accuracy. In this work, the residual convolutional neural network with autoencoder based attention (RCNN-AA) is proposed for PolSAR image classification. The scaled difference of the convolutional autoencoder with the original input patch is used as the weight, which contains information about the fine spatial features. Multiplication of this normalized difference in the input patch provides the attention feature maps that can be concatenated with the original input and used as input of the RCNN. An ablation study is done, and also, the proposed RCNN-AA model is compared to some deep learning based models. The results show preference of the RCNN-AA with respect to the competitors.
Implicit information exploration techniques are of great importance for the restoration and conservation of cultural relics. At present, the hyperspectral image analysis technique is one of the main methods to extract...
详细信息
ISBN:
(纸本)9781510664494;9781510664500
Implicit information exploration techniques are of great importance for the restoration and conservation of cultural relics. At present, the hyperspectral image analysis technique is one of the main methods to extract hidden information, which mainly contains two analysis methods such as principal component analysis (PCA) and minimum noise fraction rotation (MNF), both of which have achieved certain information extraction effects. In recent years, with the development of artificial intelligence, deep learning, and other technologies, nonlinear methods such as neural networks are expected to further improve the effect of implicit information mining. Therefore, this paper is oriented to the problem of extracting hidden information from pottery artifacts and tries to study and explore the hidden information mining method based on deep neural networks, expecting to obtain more stable and richer hidden information. In this paper, an auto-encoder-based implied information mining method is proposed first, and the auto-encoder (AE) framework achieves good performance in feature learning by automatically learning low-dimensional embedding and reconstructing data. However, during the experiments, it is found that some important detailed information (e.g., implicit information) is often lost in the reconstruction process because the traditional autoencoder network only focuses more on the pixel-level reconstruction loss and ignores the overall distribution. Therefore, this paper further proposes a multi-scale convolutional autoencoder network (MSCAE). It constructs a multi-scale convolutional module based on the traditional AE and designs a cyclic consistency loss in addition to the reconstruction loss, to reduce the loss of detailed information in the reconstruction process and improve the implicit information mining effect. In the experiments, we find that the proposed method can achieve effective implied information mining by extracting implied information from cocoon-shape
This paper presents a hardware-accelerated autoencoder (AE) for wireless communication using a Short-Time-Fourier-Transform Assisted Convolutional Neural Network (STFT-CNN-AE). The design aims to reduce the autoencode...
详细信息
ISBN:
(纸本)9798350348606;9783981926385
This paper presents a hardware-accelerated autoencoder (AE) for wireless communication using a Short-Time-Fourier-Transform Assisted Convolutional Neural Network (STFT-CNN-AE). The design aims to reduce the autoencoder's resource requirements and power dissipation while maintaining its performance even in low Signal-to-Noise Ratio (SNR) wireless channels. The STFT-CNN-AE was implemented and tested on a Zynq UltraScale+ FPGA platform. Prototype measurements show that the STFT-CNN-AE achieves 3.5 times higher throughput at 2.6 times faster frequency, consumes 59% less power, and requires 76% fewer hardware resources (LUT and DSP) compared to a prior Multi-Layer Perceptron-based AE (MLP-AE). These improvements were achieved while maintaining comparable performance in low SNR (<7.5dB) channels.
Background: Skin feature tracking enables quantification of human motion in an explainable way, making it suitable for clinical assessments. Accuracy is crucial, but no study has investigated state-of-the-art deep neu...
详细信息
ISBN:
(纸本)9798350360875;9798350360868
Background: Skin feature tracking enables quantification of human motion in an explainable way, making it suitable for clinical assessments. Accuracy is crucial, but no study has investigated state-of-the-art deep neural network-based point tracking models such as Cotracker. Cotracker jointly tracks points and has been shown to have better 3-pixel accuracy than five other state-of-the-art deep learning methods on the two most commonly used datasets for evaluation of single target point tracking. In 2021, Chang and Nordling introduced the Deep Feature Encoder (DFE) and demonstrated skin feature tracking so accurate that the errors cannot be excluded to stem from the manual labeling of the videos based on a chi(2)-test. Problem: How accurately can different methods track skin features and how to avoid the intrinsic weaknesses of the methods? Methods: We use videos of the Unified Parkinson's Disease Rating Scale postural tremor test recorded at two hospitals for benchmarking. DFE utilizes the encoder part of an autoencoder consisting of a five-layer convolutional neural network trained to reproduce skin crops without supervision. The residual squared error of the latent features of the encoder is then compared with crops to obtain a predicted position. We also propose Cotracker-DFE, using Cotracker to obtain an approximate position and subsequently cropping a small area that is fed to DFE to obtain a position predicted with a lower mean pixel error. Results: The mean Euclidean distance errors of Cotracker, DFE, and Cotracker-DFE are 1.2, 0.8, and 0.8 pixels, respectively. DFE requires time-consuming computations, making it 35 times slower than Cotracker. Conclusion: The old school DFE provided more accurate skin feature tracking, while combining DFE with Cotracker provides the best overall performance, circumventing the lack of labeled data and computational resources required to fine-tune Cotracker.
With the rapid development of the Internet, various network invasive behaviors are increasing rapidly. This seriously threatens the economic development of individuals, enterprises, and society. Network intrusion dete...
详细信息
ISBN:
(纸本)9798350373981;9798350373974
With the rapid development of the Internet, various network invasive behaviors are increasing rapidly. This seriously threatens the economic development of individuals, enterprises, and society. Network intrusion detection is important in network security systems, which can be regarded as a classification problem. It aims to distinguish between the specific categories of various network behaviors and determine whether the behavior belongs to network intrusion. However, network intrusions present a diverse and fast-changing trend, making categorizing difficult. Due to feature redundancy, uneven distribution of sample numbers, and inefficient parameter optimization, traditional rule-based approaches fail to achieve satisfying classification accuracy. This work proposes a multi-classification intrusion detection model based on Stacked Sparse Shrink autoencoder (SSSAE), Genetic Simulated annealing-based particle swarm optimization optimized Tabnet classifier (GS-Tabnet), and Decision Fusion (DF), called for SGTD short. Among them, SSSAE extracts multiple feature sets from the input data. Then GS-Tabnet trains a classifier for each feature set. Finally, the decision fusion fuses the results from these classifiers to obtain the final classification result. SGTD is compared with eight multi-classification benchmark models, and its intrusion detection accuracy is superior to its peers.
This work addresses the challenge of including the spatial dimension into the autoencoder models for lossy compression of different spatially independent and unknown hyperspectral datasets acquired by space-borne hype...
详细信息
ISBN:
(纸本)9798350360332;9798350360325
This work addresses the challenge of including the spatial dimension into the autoencoder models for lossy compression of different spatially independent and unknown hyperspectral datasets acquired by space-borne hyperspectral sensors. We propose two different 3D-Hybrid Convolutional autoencoder models with increased compression rates compared to 1D methods that can compress and reconstruct hyperspectral data with arbitrary spectral dimensionality. The architecture of the first 3D-Hybrid model consists of the A1D-CAE in combination with the 2D-CAE. The second 3D-Hybrid model includes the adaptive 1D-CAE and a 3D-CAE. The evaluation of the reconstruction accuracy is measured by comparing the spectral angle and the peak signal-to-noise ratio between the original and the reconstructed data and structural similarity index measure. We show the high transferability and generalizability of our 3D-Hybrid models on different PRISMA datasets. The 3D-Hybrid model is compared with the SSCNet(2D) based on a 2D-CAE and a 3D-CAE model. The findings of this study contribute to understanding the strengths and limitations of machine learning-based compression methods for jointly compressing spectral and spatial information.
暂无评论