Capsule Neural Network Research Articles

Hyb4mC: a hybrid DNA2vec-based model for DNA N4-methylcytosine sites prediction

BackgroundDNA N4-methylcytosine is part of the restrictive modification system, which works by regulating some biological processes, for example, the initiation of DNA replication, mismatch repair and inactivation of transposon. However, using experimental methods to detect 4mC sites is time-consuming and expensive. Besides, considering the huge differences in the number of 4mC samples among different species, it is challenging to achieve a robust multi-species 4mC site prediction performance. Hence, it is of great significance to develop effective computational tools to identify 4mC sites.ResultsThis work proposes a flexible deep learning-based framework to predict 4mC sites, called Hyb4mC. Hyb4mC adopts the DNA2vec method for sequence embedding, which captures more efficient and comprehensive information compared with the sequence-based feature method. Then, two different subnets are used for further analysis: Hyb_Caps and Hyb_Conv. Hyb_Caps is composed of a capsule neural network and can generalize from fewer samples. Hyb_Conv combines the attention mechanism with a text convolutional neural network for further feature learning.ConclusionsExtensive benchmark tests have shown that Hyb4mC can significantly enhance the performance of predicting 4mC sites compared with the recently proposed methods.

Intelligent Classification of Japonica Rice Growth Duration (GD) Based on CapsNets.

Rice cultivation in cold regions of China is mainly distributed in Heilongjiang Province, where the growing season of rice is susceptible to low temperature and cold damage. Choosing and planting rice varieties with suitable GD according to the accumulated temperate zone is an important measure to prevent low temperature and cold damage. However, the traditional identification method of rice GD requires lots of field investigations, which are time consuming and susceptible to environmental interference. Therefore, an efficient, accurate, and intelligent identification method is urgently needed. In response to this problem, we took seven rice varieties suitable for three accumulated temperature zones in Heilongjiang Province as the research objects, and we carried out research on the identification of japonica rice GD based on Raman spectroscopy and capsule neural networks (CapsNets). The data preprocessing stage used a variety of methods (signal.filtfilt, difference, segmentation, and superposition) to process Raman spectral data to complete the fusion of local features and global features and data dimension transformation. A CapsNets containing three neuron layers (one convolutional layer and two capsule layers) and a dynamic routing protocol was constructed and implemented in Python. After training 160 epochs on the CapsNets, the model achieved 89% and 93% accuracy on the training and test datasets, respectively. The results showed that Raman spectroscopy combined with CapsNets can provide an efficient and accurate intelligent identification method for the classification and identification of rice GD in Heilongjiang Province.

Evaluation and Prediction of Landslide Susceptibility in Yichang Section of Yangtze River Basin Based on Integrated Deep Learning Algorithm

Landslide susceptibility evaluation (LSE) refers to the probability of landslide occurrence in a region under a specific geological environment and trigger conditions, which is crucial to preventing and controlling landslide risk. The mainstream of the Yangtze River in Yichang City belongs to the largest basin in the Three Gorges Reservoir area and is prone to landslides. Affected by global climate change, seismic activity, and accelerated urbanization, geological disasters such as landslide collapses and debris flows in the study area have increased significantly. Therefore, it is urgent to carry out the LSE in the Yichang section of the Yangtze River Basin. The main results are as follows: (1) Based on historical landslide catalog, geological data, geographic data, hydrological data, remote sensing data, and other multi-source spatial-temporal big data, we construct the LSE index system; (2) In this paper, unsupervised Deep Embedding Clustering (DEC) algorithm and deep integration network (Capsule Neural Network based on SENet: SE-CapNet) are used for the first time to participate in non-landslide sample selection, and LSE in the study area and the accuracy of the algorithm is 96.29; (3) Based on the constructed sensitivity model and rainfall forecast data, the main driving mechanisms of landslides in the Yangtze River Basin were revealed. In this paper, the study area’s mid-long term LSE prediction and trend analysis are carried out. (4) The complete results show that the method has good performance and high precision, providing a reference for subsequent LSE, landslide susceptibility prediction (LSP), and change rule research, and providing a scientific basis for landslide disaster prevention.

Multikernel Capsule Network for Schizophrenia Identification.

Schizophrenia seriously affects the quality of life. To date, both simple (e.g., linear discriminant analysis) and complex (e.g., deep neural network) machine-learning methods have been utilized to identify schizophrenia based on functional connectivity features. The existing simple methods need two separate steps (i.e., feature extraction and classification) to achieve the identification, which disables simultaneous tuning for the best feature extraction and classifier training. The complex methods integrate two steps and can be simultaneously tuned to achieve optimal performance, but these methods require a much larger amount of data for model training. To overcome the aforementioned drawbacks, we proposed a multikernel capsule network (MKCapsnet), which was developed by considering the brain anatomical structure. Kernels were set to match partition sizes of the brain anatomical structure in order to capture interregional connectivities at the varying scales. With the inspiration of the widely used dropout strategy in deep learning, we developed capsule dropout in the capsule layer to prevent overfitting of the model. The comparison results showed that the proposed method outperformed the state-of-the-art methods. Besides, we compared performances using different parameters and illustrated the routing process to reveal characteristics of the proposed method. MKCapsnet is promising for schizophrenia identification. Our study first utilized the capsule neural network for analyzing functional connectivity of magnetic resonance imaging (MRI) and proposed a novel multikernel capsule structure with the consideration of brain anatomical parcellation, which could be a new way to reveal brain mechanisms. In addition, we provided useful information in the parameter setting, which is informative for further studies using a capsule network for other neurophysiological signal classification.

Multiclass Classification of Chest X-Ray Images for the Prediction of COVID-19 Using Capsule Network.

It is critical to establish a reliable method for detecting people infected with COVID-19 since the pandemic has numerous harmful consequences worldwide. If the patient is infected with COVID-19, a chest X-ray can be used to determine this. In this work, an X-ray showing a COVID-19 infection is classified by the capsule neural network model we trained to recognise. 6310 chest X-ray pictures were used to train the models, separated into three categories: normal, pneumonia, and COVID-19. This work is considered an improved deep learning model for the classification of COVID-19 disease through X-ray images. Viewpoint invariance, fewer parameters, and better generalisation are some of the advantages of CapsNet compared with the classic convolutional neural network (CNN) models. The proposed model has achieved an accuracy greater than 95% during the model's training, which is better than the other state-of-the-art algorithms. Furthermore, to aid in detecting COVID-19 in a chest X-ray, the model could provide extra information.

A Novel Capsule Convolutional Neural Network with Attention Mechanism for High-Voltage Circuit Breaker Fault Diagnosis

The traditional fault diagnosis method of high-voltage circuit breakers (HVCBs) based on deep learning largely depends on the huge data size. When the data sample size is insufficient, it is difficult to meet the needs of diagnosis performance. To improve the diagnostic performance, a one-dimensional attention convolutional capsule neural network is proposed in this paper. First, to take advantage of the time fine-grained information of the vibration signals, a one-dimensional convolutional neural network with a wide convolution kernel is designed, thus avoiding the tedious manual feature extraction of the original data. Then, an attention mechanism is added in the pooling layer to change the weight allocation of vibration signal fragments and increase its feature extraction capability. Finally, a capsule layer is introduced between the convolution layer and the fully connected layer to represent feature vectorization, which ensures the integrity of fault feature extraction. The classification structure digital capsule is obtained by the feature transfer of dynamic routing. The experiment results show that the combination of the attention mechanism and capsule layer is feasible and the accuracy can reach 94.7% when the number of samples is small. Compared with the traditional method, the proposed method still has high diagnostic accuracy and anti-noise ability in the case of small sample.

Twitter sentiment analysis using ensemble based deep learning model towards COVID-19 in India and European countries.

As of November 2021, more than 24.80 crore people are diagnosed with the coronavirus in that around 50.20 lakhs people lost their lives, because of this infectious disease. By understanding the people's sentiment's expressed in their social media (Facebook, Twitter, Instagram etc.) helps their governments in controlling, monitoring, and eradicating the coronavirus. Compared to other social media's, the twitter data are indispensable in the extraction of useful awareness information related to any crisis. In this article, a sentiment analysis model is proposed to analyze the real time tweets, which are related to coronavirus. Initially, around 3100 Indian and European people's tweets are collected between the time period of 23.03.2020 to 01.11.2021. Next, the data pre-processing and exploratory investigation are accomplished for better understanding of the collected data. Further, the feature extraction is performed using Term Frequency-Inverse Document Frequency (TF-IDF), GloVe, pre-trained Word2Vec, and fast text embedding's. The obtained feature vectors are fed to the ensemble classifier (Gated Recurrent Unit (GRU) and Capsule Neural Network (CapsNet)) for classifying the user's sentiment's as anger, sad, joy, and fear. The obtained experimental outcomes showed that the proposed model achieved 97.28% and 95.20% of prediction accuracy in classifying the both Indian and European people's sentiments.

A wavelet-based capsule neural network for ECG biometric identification

Electrocardiogram (ECG) signals have received a high level of attention from the biometric research community due to their unique nature for each person, which makes them suitable for developing accurate and reliable human identification systems. Although most existing ECG-based biometric recognition methods have received prominent results, several consecutive heartbeat segments are used in their approaches to achieve high accuracy, which is challenging to apply in biometric systems deployed in real-world applications. This paper proposes a new approach for human identification via ECG, based on a combination of Continuous Wavelet Transform (CWT), Discrete Wavelet Transform (DWT) along with a novel kind of deep learning technique known as Capsule network. The CWT is used to transform a single heartbeat signal into the time–frequency domain, and the DWT is adopted to extract spectral information of 2D frequency-time scalogram images to further improve accuracy. The discrete wavelet coefficients are then used as input to the capsule network promoting the recognition performance due to its high learning capacities. To support real-life practicality of ECG biometric identification system, the effectiveness and efficiency of our approach were evaluated over four databases that include normal and abnormal ECG records: PTB Diagnosis ECG (PTB), MIT-BIH Arrhythmia, MIT-BIH Normal Sinus Rhythm (NSRDB), and the MIT-BIH ST Change (STDB) databases. Experimental results demonstrate that our proposed method was able to achieve high identification accuracies and outperforming other state-of-the-art methods, by achieving an accuracy of 99.5%, 98.1%, 98.2%, and 100% on the PTB, MIT-BIH arrhythmia, STDB, and NSRDB respectively. Furthermore, the approach showed very good generalization ability since the training and test sets were completely different, which demonstrates the feasibility to promote the application of our approach in practice.

Multiscale Feature Aggregation Capsule Neural Network for Hyperspectral Remote Sensing Image Classification

Models based on capsule neural network (CapsNet), a novel deep learning method, have recently made great achievements in hyperspectral remote sensing image (HSI) classification due to their excellent ability to implicitly model the spatial relationship knowledge embedded in HSIs. However, the number of labeled samples is a common bottleneck in HSI classification, limiting the performance of these deep learning models. To alleviate the problem of limited labeled samples and further explore the potential of CapsNet in the HSI classification field, this study proposes a multiscale feature aggregation capsule neural network (MS-CapsNet) based on CapsNet via the implementation of two branches that simultaneously extract spectral, local spatial, and global spatial features to integrate multiscale features and improve model robustness. Furthermore, because deep features are generally more discriminative than shallow features, two kinds of capsule residual (CapsRES) blocks based on 3D convolutional capsule (3D-ConvCaps) layers and residual connections are proposed to increase the depth of the network and solve the limited labeled sample problem in HSI classification. Moreover, a squeeze-and-excitation (SE) block is introduced in the shallow layers of MS-CapsNet to enhance its feature extraction ability. In addition, a reasonable initialization strategy that transfers parameters from two well-designed, pretrained deep convolutional capsule networks is introduced to help the model find a good set of initializing weight parameters and further improve the HSI classification accuracy of MS-CapsNet. Experimental results on four widely used HSI datasets demonstrate that the proposed method can provide results comparable to those of state-of-the-art methods.

Wind speed prediction based on multi-variable Capsnet-BILSTM-MOHHO for WPCCC

To additional understand the wind speed prediction of every wind energy facility in several geographical locations and environments within the Wind Power Centralized Control Center (WPCCC), this paper proposes a variable short wind speed prediction model of Capsule Neural Network (Capsnet) and bidirectional Long and Short Term Memory Network (BILSTM) combined with Multi-Object Harris Hawk optimization (MOHHO). The method selects four typical location wind farms within the wind farm hub and uses their historical wind speed information and multidimensional meteorologic variables as inputs to the model, then extract the spatial features of the multifaceted temporal variables by Capsnet and apply BILSTM to simulate the time dependence of the capsule layer outputs, and at last transfer the four pre-trained models to the wind farm hub victimization the transfer learning and MOHHO optimization algorithm to weight the four sets of prediction results, to get the wind speed prediction results of any wind farm underneath the jurisdiction of the centralized center. Once examination of the experimental results, it's shown that the model has high prediction accuracy.

BiGRU attention capsule neural network for persian text classification

BiGRU attention capsule neural network for persian text classification

Detection of bladder cancer with feature fusion, transfer learning and CapsNets

This paper confronts two approaches to classify bladder lesions shown in white light cystoscopy images when using small datasets: the classical one, where handcrafted-based features feed pattern recognition systems and the modern deep learning-based (DL) approach. In between, there are alternative DL models that had not received wide attention from the scientific community, even though they can be more appropriate for small datasets such as the human brain motivated capsule neural networks (CapsNets). However, CapsNets have not yet matured hence presenting lower performances than the most classic DL models. These models require higher computational resources, more computational skills from the physician and are more prone to overfitting, making them sometimes prohibitive in the routine of clinical practice. This paper shows that carefully handcrafted features used with more robust models can reach similar performances to the conventional DL-based models and deep CapsNets, making them more useful for clinical applications. Concerning feature extraction, it is proposed a new feature fusion approach for Ta and T1 bladder tumor detection by using decision fusion from multiple classifiers in a scheme known as stacking of classifiers. Three Neural Networks perform classification on three different feature sets, namely: Covariance of Color Histogram of Oriented Gradients, proposed in the ambit of this paper; Local Binary Patterns and Wavelet Coefficients taken from lower scales. Data diversity is ensured by a fourth Neural Network, which is used for decision fusion by combining the outputs of the ensemble elements to produce the classifier output. Both Feed Forward Neural Networks and Radial Basis Functions are used in the experiments. Contrarily, DL-based models extract automatically the best features at the cost of requiring huge amounts of training data, which in turn can be alleviated by using the Transfer Learning (TL) strategy. In this paper VGG16 and ResNet-34 pretrained in ImageNet were used for TL, slightly outperforming the proposed ensemble. CapsNets may overcome CNNs given their ability to deal with objects rotational invariance and spatial relationships. Therefore, they can be trained from scratch in applications using small amounts of data, which was beneficial for the current case, improving accuracy from 94.6% to 96.9%.

EdgeLoc: A Robust and Real-Time Localization System Toward Heterogeneous IoT Devices

Indoor localization has become an essential demand driven by indoor location-based services (ILBSs) for mobile users. With the rising of Internet of Things (IoT), heterogeneous smartphones and wearables have become ubiquitous. However, the ILBSs for heterogeneous IoT devices confront significant challenges, such as received signal strength (RSS) variances caused by hardware heterogeneity, multipath reflections from complex environments, and localization time restricted by computation resources. This article proposes EdgeLoc, a robust and real-time indoor localization system toward heterogeneous IoT devices to solve the above challenges. In particular, the RSS fingerprinting data of Wi-Fi is employed for localization and tackling the heterogeneity of IoT devices in twofold. First, feature-level and signal-level solutions are presented to address the random RSS variances. At the feature level, this work proposes a novel capsule neural network model to efficiently extract incremental features from RSS fingerprinting data. At the signal level, a multistep dataflow is further devised to process RSS fingerprints into image-like data, which utilizes the feature matrix to reduce absolute sensing errors introduced by hardware heterogeneity. Second, an edge-IoT framework is designed to utilize the edge server to train the deep learning model and further supports real-time localization for heterogeneous IoT devices. Extensive field experiments with over 33 600 data points are conducted to validate the effectiveness of EdgeLoc with a large-scale Wi-Fi fingerprint data set. The results show that EdgeLoc outperforms the state-of-the-art SAE-CNN method in localization accuracy by up to 14.4%, with an average error of 0.68 m and an average positioning time of 2.05 ms.

Sentiment analysis on Bangla text using extended lexicon dictionary and deep learning algorithms

Sentiment analysis (SA) is a subset of natural language processing (NLP) research. In the case of categorical weighted based dictionary with rule-based sentiment score generation, no work in SA has been done yet in Bangla language using deep learning (DL) approaches. This paper proposes DL models for SA on Bangla text using an extended lexicon data dictionary (LDD). We implement the rule-based method Bangla text sentiment score (BTSC) algorithm for extracting polarity from large texts. These polarities are then fed into the neural network along with the preprocessed text as training samples. The preprocessed texts are formatted as a vectorization of words of unique numbers of pre-trained word embedding models. Word2Vec matrix with top highest probability word is applied on embedding layer as a weighted matrix to fit the DL models. This paper also presents a remarkably detailed analysis of selective DL models with some fine tuning. The fine-tuning includes the use of drop out, optimizer regularization, learning rate, multiple layers, filters, attention mechanism, capsule layers, transformer with progressive training along with validation and testing accuracy, precision, recall and F1-score. Experimental results indicate that the proposed new long short-term memory (LSTM) models are highly accurate in performing SA tasks. For our proposed hierarchical attention based LSTM (HAN-LSTM), Dynamic routing based capsule neural network with Bi-LSTM (D-CAPSNET-Bi-LSTM) and bidirectional encoder representations from Transformers (BERT) with LSTM (BERT-LSTM) model we achieved accuracy values of 78.52%, 80.82% and 84.18% respectively.

CARD-B: A stacked ensemble learning technique for classification of encrypted network traffic

Classification of network traffic data into different applications, services, or types is critical for network service providers to monitor networks and maintain Quality of Service (QoS). With the continuous evolution of technological advancements and with the rapid increase in security and user privacy concerns, encryption techniques are commonly used. Encrypted network traffic makes it challenging for network service providers to monitor a network using classical network monitoring techniques and tools. Due to security and privacy reasons, data cannot be decrypted. The dynamics of encrypted network traffic data cannot be interpreted, and it makes the task of classifying the encrypted network traffic a major challenge. This presents Internet Service Providers with challenges as varying Quality of Service is being provided to clients along with the security and privacy concerns to monitor network traffic. Machine learning and deep learning techniques are being utilized to classify encrypted network traffic data. This paper presents an ensemble learning technique that is based on existing data pre-processing machine learning and deep learning techniques. We examine different models and identify the best and relevant statistical features from encrypted network traffic for the classification of non-VPN encrypted network traffic data. We performed multiple experiments that led us to developing an ensemble learning model based on the existing deep learning and machine learning models for the classification of non-VPN encrypted network traffic data. The proposed solution (named as CARD-B) is composed of Capsule Neural Networks, Artificial Neural Networks, Random Forest, Decision Trees, along with Boosting techniques such as Adaptive Boosting and Extreme Gradient Boosting. The techniques are stacked using Random Forest Classifier. The result of the experiment shows that the proposed model achieved an overall accuracy of 96% and an AUC of 98% using different possible extracted statistical features.

Deep Learning Approaches for Robust Time of Arrival Estimation in Acoustic Emission Monitoring.

In this work, different types of artificial neural networks are investigated for the estimation of the time of arrival (ToA) in acoustic emission (AE) signals. In particular, convolutional neural network (CNN) models and a novel capsule neural network are proposed in place of standard statistical strategies which cannot handle, with enough robustness, very noisy scenarios and, thus, cannot be sufficiently reliable when the signal statistics are perturbed by local drifts or outliers. This concept was validated with two experiments: the pure ToA identification capability was firstly assessed on synthetic signals for which a ground truth is available, showing a 10× gain in accuracy when compared to the classical Akaike information criterion (AIC). Then, the same models were tested via experimental data acquired in the framework of a localization problem to identify targets with known coordinates on a square aluminum plate, demonstrating an overreaching precision under significant noise levels.

Emotion Recognition with Capsule Neural Network

For human-machine communication to be as effective as human-to-human communication, research on speech emotion recognition is essential. Among the models and the classifiers used to recognize emotions, neural networks appear to be promising due to the network’s ability to learn and the diversity in configuration. Following the convolutional neural network, a capsule neural network (CapsNet) with inputs and outputs that are not scalar quantities but vectors allows the network to determine the part-whole relationships that are specific 6 for an object. This paper performs speech emotion recognition based on CapsNet. The corpora for speech emotion recognition have been augmented by adding white noise and changing voices. The feature parameters of the recognition system input are mel spectrum images along with the characteristics of the sound source, vocal tract and prosody. For the German emotional corpus EMO-DB, the average accuracy score for 4 emotions, neutral, boredom, anger and happiness, is 99.69%. For Vietnamese emotional corpus BKEmo, this score is 94.23% for 4 emotions, neutral, sadness, anger and happiness. The accuracy score is highest when combining all the above feature parameters, and this score increases significantly when combining mel spectrum images with the features directly related to the fundamental frequency.

Discriminating Earthquakes From Quarry Blasts Using Capsule Neural Network

Discrimination between earthquakes and quarry blasts is crucial for precise seismic analysis, e.g., seismic hazard mitigation, earthquake cataloging, etc. However, the discrimination process is challenging due to the similarity of waveforms between the local earthquakes and quarry blasts. We propose to use the scalogram and the capsule neural network to distinguish between earthquakes and quarry blasts. First, we obtain the scalogram for 60s 3-channel waveforms, where we extract 10s before and 50s after the first arrival time of the seismic event. Secondly, we utilize the capsule neural network to extract the important information from the input scalogram which leads to robust classification performance. The proposed capsule neural network consists of the convolutional layer, primary capsule layer, and digit caps layer. The convolutional layer extracts the important information from the input data, and the primary capsule layer extracts the spatial relationship between different feature maps. Thirdly, we use the dynamic routing process to connect the primary capsule to the digit caps layer. We train and test the proposed capsule network using a small and unbalanced dataset which is recorded by the Egyptian Seismic Network (ENSN) in the Red Sea and the surrounded area in Egypt. Accordingly, the proposed method achieves a test accuracy of 96.08%. The proposed method is compared to the benchmark methods, i.e., convolutional neural network (CNN), AlexNet, VGG, and ResNet networks, and demonstrated to outperform all of the competing methods. Finally, we apply the proposed method to classify real-time seismic events and obtain promising results.

Exploring Entropy Pruning Coupled with Capsule Neural Network(Capsnet) for Leaf Disease Classification

Exploring Entropy Pruning Coupled with Capsule Neural Network(Capsnet) for Leaf Disease Classification

Overview of Capsule Neural Networks

<p>As a vector transmission network structure, the capsule neural network has been one of the research hotspots in deep learning since it was proposed in 2017. In this paper, the latest research progress of capsule networks is analyzed and summarized. Firstly, we summarize the shortcomings of convolutional neural networks and introduce the basic concept of capsule network. Secondly, we analyze and summarize the improvements in the dynamic routing mechanism and network structure of the capsule network in recent years and the combination of the capsule network with other network structures. Finally, we compile the applications of capsule network in many fields, including computer vision, natural language, and speech processing. Our purpose in writing this article is to provide methods and means that can be used for reference in the research and practical applications of capsule networks.</p> <p> </p>