Nuclear Magnetic Resonance (NMR) spectroscopy is crucial for molecular structure analysis but poses challenges due to high dimensionality and noise. We propose a deep learning approach that combines autoencoders and C...
详细信息
ISBN:
(纸本)9798350350661;9798350350654
Nuclear Magnetic Resonance (NMR) spectroscopy is crucial for molecular structure analysis but poses challenges due to high dimensionality and noise. We propose a deep learning approach that combines autoencoders and Convolutional Neural Networks (CNNs) for improved NMR spectra classification. An autoencoder reduces the data to compact latent representations, which a CNN then classifies. Our method, evaluated on a comprehensive NMR dataset, shows significant accuracy improvements over direct CNN classification on raw spectra and other baselines. The autoencoder effectively captures essential features, enhancing the CNN's performance. This framework addresses NMR classification challenges, offering a robust tool for researchers and practitioners. NMR Classification which mainly is dealt manually for now and classified by the spectra peaks as case or control, can be now classified using our proposed framework. In this we are achieving 80.9% accuracy on NMR Spectra dataset. This framework is an improved solution for NMR spectroscopy with least human intervention.
Automatic detection of human-related anomalous events in surveillance videos is challenging, owing to unclear definition of anomalies and insufficiency of training data. Generally, the irregular human motion patterns ...
详细信息
Automatic detection of human-related anomalous events in surveillance videos is challenging, owing to unclear definition of anomalies and insufficiency of training data. Generally, the irregular human motion patterns can be regarded as human-related abnormal events. Therefore, we propose a novel method to operate directly on sequences of human skeleton graphs for discovering the normal patterns of human motion. The sequence of skeleton graphs is decomposed into two sub-components: global movement and local posture sequences. The global component is utilized to compute local component. The local component sequences are then input to our network for capturing normal spatial-temporal motion patterns of human skeleton. Our network is established on a Spatial-temporal Graph Convolutional autoencoder (ST-GCAE) and embedded with Long Short-Term Memory (LSTM) network in hidden layers for exploring the temporal cues, which is thus called Spatial-temporal Graph Convolutional autoencoder with Embedded Long Short-Term Memory Network (STGCAE-LSTM). Different from traditional autoencoder, STGCAE-LSTM owns a single-encoder-dual-decoder architecture, which is capable of reconstructing the input and predicting the unseen future simultaneously. Then, samples that deviate from normal patterns are detected as anomalies with fusion of reconstruction and prediction errors. Experimental results on four challenging datasets demonstrate advantages of our method over other state-of-the-art algorithms. (c) 2021 Elsevier B.V. All rights reserved.
Domain adaptation aims to facilitate the learning task in an unlabeled target domain by leveraging the auxiliary knowledge in a well-labeled source domain from a different distribution. Almost existing autoencoder-bas...
详细信息
Domain adaptation aims to facilitate the learning task in an unlabeled target domain by leveraging the auxiliary knowledge in a well-labeled source domain from a different distribution. Almost existing autoencoder-based domain adaptation approaches focus on learning domain-invariant representations to reduce the distribution discrepancy between source and target domains. However, there is still a weakness existing in these approaches: the class-discriminative information of the two domains may be damaged while aligning the distributions of the source and target domains, which makes the samples with different classes close to each other, leading to performance degradation. To tackle this issue, we propose a novel dual-representation autoencoder (DRAE) to learn dual-domain-invariant representations for domain adaptation. Specifically, DRAE consists of three learning phases. First, DRAE learns global representations of all source and target data to maximize the interclass distance in each domain and minimize the marginal distribution and conditional distribution of both domains simultaneously. Second, DRAE extracts local representations of instances sharing the same label in both domains to maintain class-discriminative information in each class. Finally, DRAE constructs dual representations by aligning the global and local representations with different weights. Using three text and two image datasets and 12 state-of-the-art domain adaptation methods, the extensive experiments have demonstrated the effectiveness of DRAE.
With the development of several services, traffic signal preemption is carried out for emergency tasks. For performing workflow intent, cloud systems offer unlimited virtual resources for designing the traffic signal ...
详细信息
The anomaly detection for multimode industrial process is a challenging problem, because the multiple operation modes present various main distributions of monitored variables, and the dynamic sequential characteristi...
详细信息
The anomaly detection for multimode industrial process is a challenging problem, because the multiple operation modes present various main distributions of monitored variables, and the dynamic sequential characteristics exist within each operation mode. This paper proposes an anomaly detection method based on sequence-to-sequence gated recurrent units (SGRU). First, to better model both the cross-mode trends and mode-specific sequential characteristics, a main reconstruction module and residual reconstruction module are integrated to improve the ability to represent complex process. Both modules are implemented by SGRUs. Second, a reconstruction error prediction module is designed to estimate the mean values of mode-specific reconstruction errors, which helps to determine the more reliable alarm thresholds. Third, the two anomaly indicators are utilized to represent the deviation degree of monitored variables against the normal conditions, according to the statistical errors and biases of reconstructions, respectively. The effectiveness of the proposed method is validated on simulations with multimode process, and on the practical data set collected from the Cleaning-in-Place multimode process of an aseptic beverage filling line in a real factory.
Great achievements have been made during the last decades in the field of Electrical Capacitance Tomography(ECT)image ***,there is still a need to make these image reconstruction results faster and of better ***,Deep ...
详细信息
Great achievements have been made during the last decades in the field of Electrical Capacitance Tomography(ECT)image ***,there is still a need to make these image reconstruction results faster and of better ***,Deep Learning(DL)is flourishing and is adopted in many *** DL is very good at dealing with complex nonlinear functions and it is built using several series of Artificial Neural Networks(ANNs).An ECT image reconstruction model using DNN is proposed in this *** proposed model mainly uses Residual autoencoder called(ECT_ResAE).Alarge-scale dataset of 320 k instances have been generated to train and test the proposed ECT_ResAE *** instance contains two vectors;a distinct permittivity distribution and its corresponding capacitance *** capacitance vector has been modulated to generate a 66×66 image,and represented to the ECT_ResAE as an *** scalability and practicability of the ECT_ResAE network are tested using noisy data,new samples,and experimental *** experimental results show that the proposed ECT_ResAE image reconstruction model provides accurate reconstructed *** achieved an average image Correlation Coefficient(CC)of more than 99%and an averageRelative ImageError(IE)around 8.5%.
Traditional approaches for diagnosing Alzheimer's disease (AD) such as brain imaging and cerebrospinal fluid are invasive and expensive. It is desirable to develop a useful diagnostic tool by exploiting biomark-er...
详细信息
Traditional approaches for diagnosing Alzheimer's disease (AD) such as brain imaging and cerebrospinal fluid are invasive and expensive. It is desirable to develop a useful diagnostic tool by exploiting biomark-ers obtained from peripheral tissues due to their noninvasive and easily accessible characteristics. However, the capacity of using DNA methylation data in peripheral blood for predicting AD progression is rarely known. It is also challenging to develop an efficient prediction model considering the complex and high-dimensional DNA methylation data in a longitudinal study. Here, we develop two multi-task deep autoencoders, which are based on the convolutional autoencoder and long short-term memory autoencoder to learn the compressed feature representation by jointly minimizing the reconstruction error and maximizing the prediction accuracy. By benchmarking on longitudinal DNA methylation data collected from the peripheral blood in Alzheimer's Disease Neuroimaging Initiative, we demonstrate that the proposed multi-task deep autoencoders outperform state-of-the-art machine learning approaches for both predicting AD progression and reconstructing the temporal DNA methylation profiles. In addition, the proposed multi-task deep autoencoders can predict AD progression accurately using only the histor-ical DNA methylation data and the performance is further improved by including all temporal DNA methylation data. Availability:: https://***/lichen-lab/MTAE. (c) 2022 The Author(s). Published by Elsevier B.V. on behalf of Research Network of Computational and Structural Biotechnology. This is an open access article under the CC BY-NC-ND license (http://***/licenses/by-nc-nd/4.0/).
In this article, we develop a novel fully unsupervised autoencoder-based scheme for nonlinear hyperspectral pixel unmixing. A unique approach is derived where high noise and unresponsive pixels are accounted for, by a...
详细信息
In this article, we develop a novel fully unsupervised autoencoder-based scheme for nonlinear hyperspectral pixel unmixing. A unique approach is derived where high noise and unresponsive pixels are accounted for, by a unique averaging approach based on spatially aware filters built using radial basis function (RBF) kernels. A novel technique is implemented via calculating rank-equivalent kernel covariance matrices in order to estimate the unknown number of endmembers contributing to the data. Utilization of spatial information is done via RBF-based weighted averaging, which is then followed by endmember estimation via K-means clustering. The RBF distances from the cluster centers are determined to measure the position of the mixed pixels in relation to the centers, which is utilized as a preliminary estimation of the abundances. The proposed framework is robust in the presence of unresponsive pixels, while highly versatile working with different nonlinear unmixing models. Extensive numerical tests establish the superiority of the novel approach with respect to state-of-the-art methods.
In recent decades, iris recognition is a trustworthy and important biometric model for human recognition. Criminal to commercial products, citizen confirmation and border control are few application areas. The researc...
详细信息
In recent decades, iris recognition is a trustworthy and important biometric model for human recognition. Criminal to commercial products, citizen confirmation and border control are few application areas. The research work is a deep learning based integrated model for accurate iris detection and recognition. Initially, eye images are considered from two datasets, the Chinese Academy of Sciences Institute of Automation (CASIA) and the Indian Institute of Technology (IIT) Delhi v1.0. Iris region is accurately segmented using Daugman's algorithm and Circular Hough Transform (CHT). Feature extraction is hybrid that is performed using Dual Tree Complex Wavelet Transform (DTCWT), Gabor filter, Local Binary Pattern (LBP) and Gray Level Co-occurrence Matrix (GLCM) from the segmented iris regions. A Multiobjective Artificial Bee Colony (MABC) algorithm is proposed to eliminate noisy and redundant feature vectors by estimating consistent information. In MABC algorithm, two multi-objective functions are formulated as reduction in number of features and classification error rate. The selected active feature vectors are given as input to autoencoder classification for iris recognition. The experimental outcome shows that MABC-autoencoder model obtained 99.67% and 98.73% accuracy on CASIA-Iris, and IIT Delhi v1.0 iris datasets. Performance evaluation is based on accuracy, specificity, Critical Success Index (CSI), sensitivity, Fowlkes Mallows (FM) index, and Mathews Correlation Coefficient (MCC).
With the proliferation of the Internet of Things, a large amount of multivariate time series (MTS) data is being produced daily by industrial systems, corresponding in many cases to life-critical tasks. The recent ano...
详细信息
With the proliferation of the Internet of Things, a large amount of multivariate time series (MTS) data is being produced daily by industrial systems, corresponding in many cases to life-critical tasks. The recent anomaly detection researches focus on using deep learning methods to construct a normal profile for MTS. However, without proper constraints, these methods cannot capture the dependencies and dynamics of MTS and thus fail to model the normal pattern, resulting in unsatisfactory performance. This paper proposes CAE-AD, a novel contrastive autoencoder for anomaly detection in MTS, by introducing multi -grained contrasting methods to extract normal data pattern. First, to capture the temporal dependency of series, a projection layer is employed and a novel contextual contrasting method is applied to learn the robust temporal representation. Second, the projected series is transformed into two different views by using time-domain and frequency-domain data augmentation. Last, an instance contrasting method is proposed to learn local invariant characteristics. The experimental results show that CAE-AD achieves an F1-score ranging from 0.9119 to 0.9376 on the three public datasets, outperforming the baseline methods.(c) 2022 Published by Elsevier Inc.
暂无评论