scholarly journals GhoMR: Multi-Receptive Lightweight Residual Modules for Hyperspectral Classification

Sensors ◽  
2020 ◽  
Vol 20 (23) ◽  
pp. 6823
Author(s):  
Arijit Das ◽  
Indrajit Saha ◽  
Rafał Scherer
Keyword(s):  
State Of The Art ◽  
Receptive Fields ◽  
Classification Performance ◽  
Data Sets ◽  
Spectral Bands ◽  
Average Accuracy ◽  
Complex Architectures ◽  
Complex Features ◽  
Hyperspectral Classification ◽  
Simple Network

In recent years, hyperspectral images (HSIs) have attained considerable attention in computer vision (CV) due to their wide utility in remote sensing. Unlike images with three or lesser channels, HSIs have a large number of spectral bands. Recent works demonstrate the use of modern deep learning based CV techniques like convolutional neural networks (CNNs) for analyzing HSI. CNNs have receptive fields (RFs) fueled by learnable weights, which are trained to extract useful features from images. In this work, a novel multi-receptive CNN module called GhoMR is proposed for HSI classification. GhoMR utilizes blocks containing several RFs, extracting features in a residual fashion. Each RF extracts features which are used by other RFs to extract more complex features in a hierarchical manner. However, the higher the number of RFs, the greater the associated weights, thus heavier is the network. Most complex architectures suffer from this shortcoming. To tackle this, the recently found Ghost module is used as the basic building unit. Ghost modules address the feature redundancy in CNNs by extracting only limited features and performing cheap transformations on them, thus reducing the overall parameters in the network. To test the discriminative potential of GhoMR, a simple network called GhoMR-Net is constructed using GhoMR modules, and experiments are performed on three public HSI data sets—Indian Pines, University of Pavia, and Salinas Scene. The classification performance is measured using three metrics—overall accuracy (OA), Kappa coefficient (Kappa), and average accuracy (AA). Comparisons with ten state-of-the-art architectures are shown to demonstrate the effectiveness of the method further. Although lightweight, the proposed GhoMR-Net provides comparable or better performance than other networks. The PyTorch code for this study is made available at the iamarijit/GhoMR GitHub repository.

scholarly journals On data collection time by an electronic nose

2021 ◽  
Vol 11 (6) ◽  
pp. 4767
Author(s):  
Piotr Borowik ◽  
Leszek Adamowicz ◽  
Rafał Tarakowski ◽  
Krzysztof Siwek ◽  
Tomasz Grzywacz
Keyword(s):  
Electronic Nose ◽  
Cross Validation ◽  
Model Performance ◽  
Classification Performance ◽  
Optimal Time ◽  
Classification Models ◽  
Machine Learning Classification ◽  
Average Accuracy ◽  
Complex Features ◽  
Collection Time

<p>We use electronic nose data of odor measurements to build machine learning classification models. The presented analysis focused on determining the optimal time of measurement, leading to the best model performance. We observe that the most valuable information for classification is available in data collected at the beginning of adsorption and the beginning of the desorption phase of measurement. We demonstrated that the usage of complex features extracted from the sensors’ response gives better classification performance than use as features only raw values of sensors’ response, normalized by baseline. We use a group shuffling cross-validation approach for determining the reported models’ average accuracy and standard deviation.</p>

Joint Concept Correlation and Feature-Concept Relevance Learning for Multilabel Classification

2018 ◽  
Vol 30 (2) ◽  
pp. 526-545
Author(s):  
Xiaowei Zhao ◽  
Zhigang Ma ◽  
Zhi Li ◽  
Zhihui Li
Keyword(s):  
State Of The Art ◽  
Relational Learning ◽  
Classification Performance ◽  
Data Sets ◽  
Classification Methods ◽  
Average Precision ◽  
Multilabel Classification ◽  
Multimedia Annotation ◽  
Multilabel Learning ◽  
Significant Attention

In recent years, multilabel classification has attracted significant attention in multimedia annotation. However, most of the multilabel classification methods focus only on the inherent correlations existing among multiple labels and concepts and ignore the relevance between features and the target concepts. To obtain more robust multilabel classification results, we propose a new multilabel classification method aiming to capture the correlations among multiple concepts by leveraging hypergraph that is proved to be beneficial for relational learning. Moreover, we consider mining feature-concept relevance, which is often overlooked by many multilabel learning algorithms. To better show the feature-concept relevance, we impose a sparsity constraint on the proposed method. We compare the proposed method with several other multilabel classification methods and evaluate the classification performance by mean average precision on several data sets. The experimental results show that the proposed method outperforms the state-of-the-art methods.

scholarly journals Gesture recognition system based on CNN-IndRNN and OpenBCI

2021 ◽  
Vol 336 ◽  
pp. 06003
Author(s):  
Na Wu ◽  
Hao JIN ◽  
Xiachuan Pei ◽  
Shurong Dong ◽  
Jikui Luo ◽  
...  
Keyword(s):  
Neural Network ◽  
Gesture Recognition ◽  
Recognition Accuracy ◽  
State Of The Art ◽  
Recognition System ◽  
Classification Performance ◽  
Non Invasive ◽  
Average Accuracy ◽  
Open Source Hardware ◽  
Semg Signals

Surface electromyography (sEMG), as a key technology of non-invasive muscle computer interface, is an important method of human-computer interaction. We proposed a CNN-IndRNN (Convolutional Neural Network-Independent Recurrent Neural Network) hybrid algorithm to analyse sEMG signals and classify hand gestures. Ninapro’s dataset of 10 volunteers was used to develop the model, and by using only one time-domain feature (root mean square of sEMG), an average accuracy of 87.43% on 18 gestures is achieved. The proposed algorithm obtains a state-of-the-art classification performance with a significantly reduced model. In order to verify the robustness of the CNN-IndRNN model, a compact real¬time recognition system was constructed. The system was based on open-source hardware (OpenBCI) and a custom Python-based software. Results show that the 10-subject rock-paper-scissors gesture recognition accuracy reaches 99.1%.

scholarly journals SMILE: a feature-based temporal abstraction framework for event-interval sequence classification

2020 ◽  
Author(s):  
Jonathan Rebane ◽  
Isak Karlsson ◽  
Leon Bornemann ◽  
Panagiotis Papapetrou
Keyword(s):  
Empirical Evaluation ◽  
Predictive Performance ◽  
Classification Performance ◽  
Drug Event ◽  
Data Sets ◽  
Temporal Abstraction ◽  
Current State ◽  
Temporal Intervals ◽  
Feature Based ◽  
Complex Features

AbstractIn this paper, we study the problem of classification of sequences of temporal intervals. Our main contribution is a novel framework, which we call , for extracting relevant features from interval sequences to construct classifiers. introduces the notion of utilizing random temporal abstraction features, we define as , as a means to capture information pertaining to class-discriminatory events which occur across the span of complete interval sequences. Our empirical evaluation is applied to a wide array of benchmark data sets and fourteen novel datasets for adverse drug event detection. We demonstrate how the introduction of simple sequential features, followed by progressively more complex features each improve classification performance. Importantly, this investigation demonstrates that significantly improves AUC performance over the current state-of-the-art. The investigation also reveals that the selection of underlying classification algorithm is important to achieve superior predictive performance, and how the number of features influences the performance of our framework.

Wavelet-based extended morphological profile and deep autoencoder for hyperspectral image classification

2018 ◽  
Vol 16 (03) ◽  
pp. 1850016 ◽  
Author(s):  
Huiwu Luo ◽  
Yuan Yan Tang ◽  
Robert P. Biuk-Aghai ◽  
Xu Yang ◽  
Lina Yang ◽  
...  
Keyword(s):  
Spatial Information ◽  
Hyperspectral Image ◽  
State Of The Art ◽  
Classification Performance ◽  
Fine Tuning ◽  
Spectral Information ◽  
Data Sets ◽  
Spatial Description ◽  
Hyperspectral Image Classification ◽  
High Level

In this paper, we propose a novel scheme to learn high-level representative features and conduct classification for hyperspectral image (HSI) data in an automatic fashion. The proposed method is a collaboration of a wavelet-based extended morphological profile (WTEMP) and a deep autoencoder (DAE) (“WTEMP-DAE”), with the aim of exploiting the discriminative capability of DAE when using WTEMP features as the input. Each part of WTEMP-DAE is ingenious and contributes to the final classification performance. Specifically, in WTEMP-DAE, the spatial information is extracted from the WTEMP, which is then joined with the wavelet denoised spectral information to form the spectral-spatial description of HSI data. The obtained features are fed into DAE as the original input, where the good weights and bias of the network are initialized through unsupervised pre-training. Once the pre-training is completed, the reconstruction layers are discarded and a logistic regression (LR) layer is added to the top of the network to perform supervised fine-tuning and classification. Experimental results on two real HSI data sets demonstrate that the proposed strategy improves classification performance in comparison with other state-of-the-art hand-crafted feature extractors and their combinations.

scholarly journals Co-Training Semi-Supervised Deep Learning for Sentiment Classification of MOOC Forum Posts

Symmetry ◽  
2019 ◽  
Vol 12 (1) ◽  
pp. 8
Author(s):  
Jing Chen ◽  
Jun Feng ◽  
Xia Sun ◽  
Yang Liu
Keyword(s):  
Deep Learning ◽  
Large Scale ◽  
Sample Selection ◽  
Classification Performance ◽  
Sentiment Classification ◽  
Unlabeled Data ◽  
Learning Performance ◽  
Average Accuracy ◽  
Complex Features

Sentiment classification of forum posts of massive open online courses is essential for educators to make interventions and for instructors to improve learning performance. Lacking monitoring on learners’ sentiments may lead to high dropout rates of courses. Recently, deep learning has emerged as an outstanding machine learning technique for sentiment classification, which extracts complex features automatically with rich representation capabilities. However, deep neural networks always rely on a large amount of labeled data for supervised training. Constructing large-scale labeled training datasets for sentiment classification is very laborious and time consuming. To address this problem, this paper proposes a co-training, semi-supervised deep learning model for sentiment classification, leveraging limited labeled data and massive unlabeled data simultaneously to achieve performance comparable to those methods trained on massive labeled data. To satisfy the condition of two views of co-training, we encoded texts into vectors from views of word embedding and character-based embedding independently, considering words’ external and internal information. To promote the classification performance with limited data, we propose a double-check strategy sample selection method to select samples with high confidence to augment the training set iteratively. In addition, we propose a mixed loss function both considering the labeled data with asymmetric and unlabeled data. Our proposed method achieved a 89.73% average accuracy and an 93.55% average F1-score, about 2.77% and 3.2% higher than baseline methods. Experimental results demonstrate the effectiveness of the proposed model trained on limited labeled data, which performs much better than those trained on massive labeled data.

scholarly journals Completely Automated CNN Architecture Design Based on Blocks

2020 ◽  
Author(s):  
Y Sun ◽  
Bing Xue ◽  
Mengjie Zhang ◽  
GG Yen
Keyword(s):  
Genetic Algorithm ◽  
Classification Accuracy ◽  
Domain Knowledge ◽  
State Of The Art ◽  
Classification Performance ◽  
Data Sets ◽  
Computational Resource ◽  
Problem Domain ◽  
Benchmark Data ◽  
Interested User

© 2019 IEEE. The performance of convolutional neural networks (CNNs) highly relies on their architectures. In order to design a CNN with promising performance, extensive expertise in both CNNs and the investigated problem domain is required, which is not necessarily available to every interested user. To address this problem, we propose to automatically evolve CNN architectures by using a genetic algorithm (GA) based on ResNet and DenseNet blocks. The proposed algorithm is completely automatic in designing CNN architectures. In particular, neither preprocessing before it starts nor postprocessing in terms of CNNs is needed. Furthermore, the proposed algorithm does not require users with domain knowledge on CNNs, the investigated problem, or even GAs. The proposed algorithm is evaluated on the CIFAR10 and CIFAR100 benchmark data sets against 18 state-of-the-art peer competitors. Experimental results show that the proposed algorithm outperforms the state-of-the-art CNNs hand-crafted and the CNNs designed by automatic peer competitors in terms of the classification performance and achieves a competitive classification accuracy against semiautomatic peer competitors. In addition, the proposed algorithm consumes much less computational resource than most peer competitors in finding the best CNN architectures.

scholarly journals Completely Automated CNN Architecture Design Based on Blocks

2020 ◽  
Author(s):  
Y Sun ◽  
Bing Xue ◽  
Mengjie Zhang ◽  
GG Yen
Keyword(s):  
Genetic Algorithm ◽  
Classification Accuracy ◽  
Domain Knowledge ◽  
State Of The Art ◽  
Classification Performance ◽  
Data Sets ◽  
Computational Resource ◽  
Problem Domain ◽  
Benchmark Data ◽  
Interested User

© 2019 IEEE. The performance of convolutional neural networks (CNNs) highly relies on their architectures. In order to design a CNN with promising performance, extensive expertise in both CNNs and the investigated problem domain is required, which is not necessarily available to every interested user. To address this problem, we propose to automatically evolve CNN architectures by using a genetic algorithm (GA) based on ResNet and DenseNet blocks. The proposed algorithm is completely automatic in designing CNN architectures. In particular, neither preprocessing before it starts nor postprocessing in terms of CNNs is needed. Furthermore, the proposed algorithm does not require users with domain knowledge on CNNs, the investigated problem, or even GAs. The proposed algorithm is evaluated on the CIFAR10 and CIFAR100 benchmark data sets against 18 state-of-the-art peer competitors. Experimental results show that the proposed algorithm outperforms the state-of-the-art CNNs hand-crafted and the CNNs designed by automatic peer competitors in terms of the classification performance and achieves a competitive classification accuracy against semiautomatic peer competitors. In addition, the proposed algorithm consumes much less computational resource than most peer competitors in finding the best CNN architectures.

scholarly journals DEEP NO LEARNING APPROACH FOR UNSUPERVISED CHANGE DETECTION IN HYPERSPECTRAL IMAGES

2021 ◽  
Vol V-3-2021 ◽  
pp. 311-316
Author(s):  
S. Saha ◽  
L. Kondmann ◽  
X. X. Zhu
Keyword(s):  
Change Detection ◽  
Receptive Fields ◽  
Super Resolution ◽  
Hyperspectral Images ◽  
Data Sets ◽  
Semantic Features ◽  
Spectral Bands ◽  
Deep Model ◽  
Significant Difference ◽  
Remarkable Result

Abstract. Unsupervised deep transfer-learning based change detection (CD) methods require pre-trained feature extractor that can be used to extract semantic features from the target bi-temporal scene. However, it is difficult to obtain such feature extractors for hyperspectral images. Moreover, it is not trivial to reuse the models trained with the multispectral images for the hyperspectral images due to the significant difference in number of spectral bands. While hyperspectral images show large number of spectral bands, they generally show much less spatial complexity, thus reducing the requirement of large receptive fields of convolution filters. Recent works in the computer vision have shown that even untrained networks can yield remarkable result in different tasks like super-resolution and surface reconstruction. Motivated by this, we make a bold proposition that untrained deep model, initialized with some weight initialization strategy can be used to extract useful semantic features from bi-temporal hyperspectral images. Thus, we couple an untrained network with Deep Change Vector Analysis (DCVA), a popular method for unsupervised CD, to propose an unsupervised CD method for hyperspectral images. We conduct experiments on two hyperspectral CD data sets, and the results demonstrate advantages of the proposed unsupervised method over other competitors.

TrustSVD: A Novel Trust-Based Matrix Factorization Model with User Trust and Item Ratings

2017 ◽  
Vol 7 (11) ◽  
pp. 7 ◽  
Author(s):  
K Sobha Rani
Keyword(s):  
Matrix Factorization ◽  
Social Trust ◽  
State Of The Art ◽  
Data Sets ◽  
Real World Data ◽  
Recommendation Algorithm ◽  
Active User ◽  
Factorization Model ◽  
The Social ◽  
Matrix Factorization Technique

Collaborative filtering suffers from the problems of data sparsity and cold start, which dramatically degrade recommendation performance. To help resolve these issues, we propose TrustSVD, a trust-based matrix factorization technique. By analyzing the social trust data from four real-world data sets, we conclude that not only the explicit but also the implicit influence of both ratings and trust should be taken into consideration in a recommendation model. Hence, we build on top of a state-of-the-art recommendation algorithm SVD++ which inherently involves the explicit and implicit influence of rated items, by further incorporating both the explicit and implicit influence of trusted users on the prediction of items for an active user. To our knowledge, the work reported is the first to extend SVD++ with social trust information. Experimental results on the four data sets demonstrate that our approach TrustSVD achieves better accuracy than other ten counterparts, and can better handle the concerned issues.

Sign in / Sign up

Export Citation Format

Share Document