Herein, we present a system for hyperspectral image segmentation that utilizes multiple class-based denoising autoencoders which are efficiently trained. Moreover, we present a novel hyperspectral data augmentation me...
详细信息
ISBN:
(纸本)9781538671504
Herein, we present a system for hyperspectral image segmentation that utilizes multiple class-based denoising autoencoders which are efficiently trained. Moreover, we present a novel hyperspectral data augmentation method for labelled HSI data using linear mixtures of pixels from each class, which helps the system with edge pixels which are almost always mixed pixels. Finally, we utilize a deep neural network and morphological hole-filling to provide robust image classification. Results run on the Salinas dataset verify the high performance of the proposed algorithm.
Remaining Useful Life (RUL) prediction for aircraft engines based on the available run-to-failure measurements of similar systems becomes more prevalent in Prognostic Health Management (PHM) thanks to the new advanced...
详细信息
Remaining Useful Life (RUL) prediction for aircraft engines based on the available run-to-failure measurements of similar systems becomes more prevalent in Prognostic Health Management (PHM) thanks to the new advanced methods of estimation. However, feature extraction and RUL prediction are challenging tasks, especially for data-driven prognostics. The key issue is how to design a suitable feature extractor that is able to give a raw of time-varying sensors measurements more meaningful representation to enhance prediction accuracy with low computational costs. In this paper, a new denoising Online Sequential Extreme Learning Machine (DOS-ELM) with double dynamic forgetting factors (DDFF) and Updated Selection Strategy (USS) is proposed. First, depending on the characteristics of the training data that comes from aircraft sensors, robust feature extraction using a modified denoising autoencoder (DAE) is introduced to learn important patterns from data. Then, USS is integrated to ensure that only the useful data sequences pass through the training process. Finally, OS-ELM is used to fit the non-accumulative linear degradation function of the engine and to address dynamic programming by trucking the new coming data and forgetting gradually the old ones based on the proposed DDFF. The proposed DOS-ELM is tested on the public dataset of commercial modular aeropropulsion system simulation (C-MAPSS) of a turbofan engine and compared with OS-ELM trained with ordinary autoencoder (AE), basic OS-ELM and previous works from the literature. Comparison results prove the effectiveness of the new integrated robust feature extraction scheme by showing more stability of the network responses even under random solutions.
A method to locate sound sources using an audio recording system mounted on an unmanned aerial vehicle (UAV) is proposed. The method introduces extension algorithms to apply on top of a baseline approach, which perfor...
详细信息
A method to locate sound sources using an audio recording system mounted on an unmanned aerial vehicle (UAV) is proposed. The method introduces extension algorithms to apply on top of a baseline approach, which performs localisation by estimating the peak signal-to-noise ratio (SNR) response in the time-frequency and angular spectra with the time difference of arrival information. The proposed extensions include a noise reduction and a post-processing algorithm to address the challenges in a UAV setting. The noise reduction algorithm reduces influences of UAV rotor noise on localisation performance, by scaling the SNR response using power spectral density of the UAV rotor noise, estimated using a denoising autoencoder. For the source tracking problem, an angular spectral range restricted peak search and link post-processing algorithm is also proposed to filter out incorrect location estimates along the localisation path. Experimental results show the proposed extensions yielded improvements in locating the target sound source correctly, with a 0.0064-0.175 decrease in mean haversine distance error across various UAV operating scenarios. The proposed method also shows a reduction in unexpected location estimations, with a 0.0037-0.185 decrease in the 0.75 quartile haversine distance error.
With respect to the problem of the low accuracy of traditional building energy prediction methods, this paper proposes a novel prediction method for building energy consumption, which is based on the seamless integrat...
详细信息
With respect to the problem of the low accuracy of traditional building energy prediction methods, this paper proposes a novel prediction method for building energy consumption, which is based on the seamless integration of the deep neural network and transfer reinforcement learning (DNN-TRL). The method introduces a stack denoising autoencoder to extract the deep features of the building energy consumption, and shares the hidden layer structure to transfer the common information between different building energy consumption problems. The output of the DNN model is used as the input of the Sarsa algorithm to improve the prediction performance of the target building energy consumption. To verify the performance of the DNN-TRL algorithm, based on the data recorded by American Power Balti Gas and Electric Power Company, and compared with Sarsa, ADE-BPNN, and BP-Adaboost algorithms, the experimental results show that the DNN-TRL algorithm can effectively improve the prediction accuracy of the building energy consumption.
BackgroundThe pathogenesis of asthma is a complex process involving multiple genes and pathways. Identifying biomarkers from asthma datasets, especially those that include heterogeneous subpopulations, is challenging....
详细信息
BackgroundThe pathogenesis of asthma is a complex process involving multiple genes and pathways. Identifying biomarkers from asthma datasets, especially those that include heterogeneous subpopulations, is challenging. Potentially, autoencoders provide ideal frameworks for such tasks as they can embed complex, noisy high-dimensional gene expression data into a low-dimensional latent space in an unsupervised fashion, enabling us to extract distinguishing features from expression ***, we developed a framework combining a denoising autoencoder and a supervised learning classifier to identify gene signatures related to asthma severity. Using the trained autoencoder with 50 hidden units, we found that hierarchical clustering on the low-dimensional embedding corresponds well with previously defined and clinically relevant clusters of patients. Moreover, each hidden unit has contributions from each of the genes, and pathway analysis of these contributions shows that the hidden units are significantly enriched in known asthma-related pathways. We then used genes that contribute most to the hidden units to develop a secondary random-forest classifier for directly predicting asthma severity. The feature importance metric from this classifier identified a signature based on 50 key genes, which are associated with severity. Furthermore, we can use these key genes to successfully estimate FEV1/FVC ratios across patients, via support-vector-machine *** found that the denoising autoencoder framework can extract meaningful patterns corresponding to functional gene groups and patient clusters from the gene expression of asthma patients.
In this paper we propose to embed SMPL within a deep-based model to accurately estimate 3D pose and shape from a still RGB image. We use CNN-based 3D joint predictions as an intermediate representation to regress SMPL...
详细信息
In this paper we propose to embed SMPL within a deep-based model to accurately estimate 3D pose and shape from a still RGB image. We use CNN-based 3D joint predictions as an intermediate representation to regress SMPL pose and shape parameters. Later, 3D joints are reconstructed again in the SMPL output. This module can be seen as an autoencoder where the encoder is a deep neural network and the decoder is SMPL model. We refer to this as SMPL reverse (SMPLR). By implementing SMPLR as an encoder-decoder we avoid the need of complex constraints on pose and shape. Furthermore, given that in-the-wild datasets usually lack accurate 3D annotations, it is desirable to lift 2D joints to 3D without pairing 3D annotations with RGB images. Therefore, we also propose a denoising autoencoder (DAE) module between CNN and SMPLR, able to lift 2D joints to 3D and partially recover from structured error. We evaluate our method on SURREAL and Human3.6M datasets, showing improvement over SMPL-based state-of-the-art alternatives by about 4 and 12 mm, respectively. (C) 2020 Elsevier Ltd. All rights reserved.
In complex underwater environments, the single mode of a single sensor cannot meet the precision requirement of object identification, and multisource fusion is currently the mainstream research approach. Deep canonic...
详细信息
In complex underwater environments, the single mode of a single sensor cannot meet the precision requirement of object identification, and multisource fusion is currently the mainstream research approach. Deep canonical correlation analysis is an efficient feature fusion method but suffers from problems such as not strong scalability and low efficiency. Therefore, an improved deep canonical correlation analysis fusion method is proposed for underwater multisource sensor data containing noise. First, a denoising autoencoder is used for denoising and to reduce the data dimension to extract new feature expressions of raw data. Second, given that underwater acoustic data can be characterized as 1-dimensional time series, a 1-dimensional convolutional neural network is used to improve the deep canonical correlation analysis model, and multilayer convolution and pooling are implemented to decrease the number of parameters and increase the efficiency. To improve the scalability and robustness of the model, a stochastic decorrelation loss function is used to optimize the objective function, which reduces the algorithm complexity from O(n(3)) to O(n(2)). The comparison experiment of the proposed algorithm and other typical algorithms on MNIST containing noise and underwater multisource data in different scenes shows that the proposed algorithm is superior to others regardless of the efficiency or precision of target classification.
Despite of remarkable progress on deep learning, its hardware implementation beyond deep learning acceleration is still behind the software deep learning due in part to lack of hardware-compatible learning algorithm. ...
详细信息
ISBN:
(纸本)9781728103976
Despite of remarkable progress on deep learning, its hardware implementation beyond deep learning acceleration is still behind the software deep learning due in part to lack of hardware-compatible learning algorithm. In this paper, a learning method called the stochastic learning with backpropagation (SLBP) algorithm was proposed. The network of concern consists of ternary synaptic weight, favorable to be implemented in a resistance-based crossbar array. Every training epoch, the SLBP algorithm evaluates weight update probability at which the corresponding weight is updated in a stochastic manner. The algorithm was used to train a denoising autoencoder, which identified the successful reduction in noise (increase in peak signal-to-noise ratio by approximately 68%). Notably, the SLBP algorithm achieves an 86% reduction in memory usage compared with a real-valued autoencoder trained using a backpropagation algorithm.
This study aims to improve the performance of speaker verification system when no labeled out-of-domain data is available. An autoencoder-based semi-supervised curriculum learning scheme is proposed to automatically c...
详细信息
This study aims to improve the performance of speaker verification system when no labeled out-of-domain data is available. An autoencoder-based semi-supervised curriculum learning scheme is proposed to automatically cluster unlabeled data and iteratively update the corpus during training. This new training scheme allows us to (1) progressively expand the size of training corpus by utilizing unlabeled data and correcting previous labels at run-time;and (2) improve robustness when generalizing to multiple conditions, such as out-of-domain and text-independent speaker verification tasks. It is also discovered that a denoising autoencoder can significantly enhance the clustering accuracy when it is trained on carefully-selected subset of speakers. Our experimental results show a relative reduction of 30% - 50% in EER compared to the baseline.
暂无评论