Open AccessPosted Content
1D Convolutional Neural Networks and Applications: A Survey
TL;DR: A comprehensive review of the general architecture and principals of 1D CNNs along with their major engineering applications, especially focused on the recent progress in this field, is presented in this paper, where the benchmark datasets and the principal 1D convolutional neural network software used in those applications are also publically shared in a dedicated website.
read more
Abstract: During the last decade, Convolutional Neural Networks (CNNs) have become the de facto standard for various Computer Vision and Machine Learning operations. CNNs are feed-forward Artificial Neural Networks (ANNs) with alternating convolutional and subsampling layers. Deep 2D CNNs with many hidden layers and millions of parameters have the ability to learn complex objects and patterns providing that they can be trained on a massive size visual database with ground-truth labels. With a proper training, this unique ability makes them the primary tool for various engineering applications for 2D signals such as images and video frames. Yet, this may not be a viable option in numerous applications over 1D signals especially when the training data is scarce or application-specific. To address this issue, 1D CNNs have recently been proposed and immediately achieved the state-of-the-art performance levels in several applications such as personalized biomedical data classification and early diagnosis, structural health monitoring, anomaly detection and identification in power electronics and motor-fault detection. Another major advantage is that a real-time and low-cost hardware implementation is feasible due to the simple and compact configuration of 1D CNNs that perform only 1D convolutions (scalar multiplications and additions). This paper presents a comprehensive review of the general architecture and principals of 1D CNNs along with their major engineering applications, especially focused on the recent progress in this field. Their state-of-the-art performance is highlighted concluding with their unique properties. The benchmark datasets and the principal 1D CNN software used in those applications are also publically shared in a dedicated website.
read more
Chat with Paper
AI Agents for this Paper
Find similar papers on Google Scholar, PubMed and Arxiv
Write a critical review of this paper
Analyze citations of this paper to find unaddressed research gaps
Figures
![Figure 4: The configuration of the first deep CNN, the “AlexNet” [28]. There are 5 convolutional layers and 3 maxpooling layers following by three (two hidden and one output) fully-connected (dense) layers. The numbers of both convolution and fully-connected layers are significantly higher than the LeNet. The neurons at the output layer use softmax loss of the network predictions for 1000 classes.](/figures/figure4-1-28molq2vx5l3.png)
Figure 4: The configuration of the first deep CNN, the “AlexNet” [28]. There are 5 convolutional layers and 3 maxpooling layers following by three (two hidden and one output) fully-connected (dense) layers. The numbers of both convolution and fully-connected layers are significantly higher than the LeNet. The neurons at the output layer use softmax loss of the network predictions for 1000 classes. ![Figure 10: The creation of the training dataset as proposed in [49] for an arbitrary user (Person-X) using the real Nbeats. The dataset is then used to train the dedicated 1D CNN for the user.](/figures/figure10-1-25wsqt1ykbbn.png)
Figure 10: The creation of the training dataset as proposed in [49] for an arbitrary user (Person-X) using the real Nbeats. The dataset is then used to train the dedicated 1D CNN for the user. ![Figure 15: Implementation (left) and configuration (right) of the 4-cell MMC circuit [54].](/figures/figure15-1-1ygy2luu42d2.png)
Figure 15: Implementation (left) and configuration (right) of the 4-cell MMC circuit [54]. ![Figure 16: The main blocks of the proposed system in [56] and the offline training of the compact 1D CNN.](/figures/figure16-1-3vilgq2s6ypp.png)
Figure 16: The main blocks of the proposed system in [56] and the offline training of the compact 1D CNN. ![Figure 1: A biological neuron (left) with the direction of the signal flow and a synapse (right) [7].](/figures/figure1-1-62sk4ss53d5t.png)
Figure 1: A biological neuron (left) with the direction of the signal flow and a synapse (right) [7]. 
Figure 5: A sample 1D CNN configuration with 3 CNN and 2 MLP layers.
![Figure 4: The configuration of the first deep CNN, the “AlexNet” [28]. There are 5 convolutional layers and 3 maxpooling layers following by three (two hidden and one output) fully-connected (dense) layers. The numbers of both convolution and fully-connected layers are significantly higher than the LeNet. The neurons at the output layer use softmax loss of the network predictions for 1000 classes.](/figures/figure4-1-28molq2vx5l3.webp)
![Figure 10: The creation of the training dataset as proposed in [49] for an arbitrary user (Person-X) using the real Nbeats. The dataset is then used to train the dedicated 1D CNN for the user.](/figures/figure10-1-25wsqt1ykbbn.webp)
![Figure 15: Implementation (left) and configuration (right) of the 4-cell MMC circuit [54].](/figures/figure15-1-1ygy2luu42d2.webp)
![Figure 16: The main blocks of the proposed system in [56] and the offline training of the compact 1D CNN.](/figures/figure16-1-3vilgq2s6ypp.webp)
![Figure 1: A biological neuron (left) with the direction of the signal flow and a synapse (right) [7].](/figures/figure1-1-62sk4ss53d5t.webp)
Citations
Chemometrics; what do we mean with it, and what do we want from it?
TL;DR: My views will be focussed on the role of chemometrics in chemistry, on how I see statistics as partly consistent, and partly inconsistent with the framework of chemistry, and where the authors should aim for the future.
318
A systematic review of convolutional neural network-based structural condition assessment techniques
TL;DR: A detailed literature review of existing CNN-based techniques in the context of infrastructure monitoring and maintenance and a brief conclusion on potential future research directions of CNN in structural condition assessment is presented.
285
A deep convolutional neural network model for rapid prediction of fluvial flood inundation
TL;DR: An innovative modelling approach based on a deep convolutional neural network (CNN) method for rapid prediction of fluvial flood inundation and shows that the CNN model outperforms SVR by a large margin.
243
A Multitier Deep Learning Model for Arrhythmia Detection
TL;DR: A deep neural network strategy is presented to ameliorate the difficulties faced in ECG-based CVD diagnosis and treatment and suggests that the proposed model could serve as an analytic module to alert users and/or medical experts when anomalies are detected.
Detection of Epileptic Seizure Using Pretrained Deep Convolutional Neural Network and Transfer Learning.
Hidir Selcuk Nogay,Hojjat Adeli +1 more
TL;DR: In this article, an end-to-end machine learning model is presented for detection of epileptic seizure using the pretrained deep two-dimensional convolutional neural network (CNN) and the concept of transfer learning.
150
References
Real-Time Motor Fault Detection by 1-D Convolutional Neural Networks
TL;DR: A fast and accurate motor condition monitoring and early fault-detection system using 1-D convolutional neural networks that has an inherent adaptive design to fuse the feature extraction and classification phases of the motor fault detection into a single learning body is proposed.
1.2K
•Proceedings Article
Unsupervised feature learning for audio classification using convolutional deep belief networks
Honglak Lee,Peter T. Pham,Yan Largman,Andrew Y. Ng +3 more
- 07 Dec 2009
TL;DR: In this paper, the authors apply convolutional deep belief networks to audio data and empirically evaluate them on various audio classification tasks and show that the learned features correspond to phones/phonemes.
A deep convolutional neural network with new training methods for bearing fault diagnosis under noisy environment and different working load
TL;DR: An end-to-end method that takes raw temporal signals as inputs and thus doesn’t need any time consuming denoising preprocessing and can achieve high accuracy when working load is changed is proposed.
1.1K
Convolutional Neural Network Based Fault Detection for Rotating Machinery
Olivier Janssens,Viktor Slavkovikj,Bram Vervisch,Kurt Stockman,Mia Loccufier,Steven Verstockt,Rik Van de Walle,Sofie Van Hoecke +7 more
TL;DR: A feature learning model for condition monitoring based on convolutional neural networks is proposed to autonomously learn useful features for bearing fault detection from the data itself and significantly outperforms the classical feature-engineering based approach which uses manually engineered features and a random forest classifier.
1.1K
Receptive fields of cells in striate cortex of very young, visually inexperienced kittens.
David H. Hubel,Torsten N. Wiesel +1 more
TL;DR: The purpose was to learn the age at which cortical cells have normal, adult-type receptive fields, and to find out whether such fields exist even in animals that have had no patterned visual stimulation.
1.1K
Related Papers (5)
Kaiming He,Xiangyu Zhang,Shaoqing Ren,Jian Sun +3 more
- 27 Jun 2016
[...]