Comparison of CNN Algorithms on Hyperspectral Image Classification in Agricultural Lands

Hsieh, Tien-Heng; Kiang, Jean‐Fu

doi:10.3390/s20061734

Cited by 75 publications

(38 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…HSI has been widely used in environmental monitoring [1], mineral exploration [2], agricultural remote sensing [3], vegetation ecology [4], ocean remote sensing [5], and other earth observation tasks. In these applications, because HSI exhibits mixed land cover categories, resulting in high intraclass variability and interclass similarity, it is a huge challenge for any classification model.…”

Section: Introductionmentioning

confidence: 99%

Hybrid Depth-Separable Residual Networks for Hyperspectral Image Classification

Zhao

Wang

et al. 2020

Complexity

View full text Add to dashboard Cite

At present, the classification of the hyperspectral image (HSI) based on the deep convolutional network has made great progress. Due to the high dimensionality of spectral features, limited samples of ground truth, and high nonlinearity of hyperspectral data, effective classification of HSI based on deep convolutional neural networks is still difficult. This paper proposes a novel deep convolutional network structure, namely, a hybrid depth-separable residual network, for HSI classification, called HDSRN. The HDSRN model organically combines 3D CNN, 2D CNN, multiresidual network ROR, and depth-separable convolutions to extract deeper abstract features. On the one hand, due to the addition of multiresidual structures and skip connections, this model can alleviate the problem of over fitting, help the backpropagation of gradients, and extract features more fully. On the other hand, the depth-separable convolutions are used to learn the spatial feature, which reduces the computational cost and alleviates the decline in accuracy. Extensive experiments on the popular HSI benchmark datasets show that the performance of the proposed network is better than that of the existing prevalent methods.

show abstract

Section: Introductionmentioning

confidence: 99%

Hybrid Depth-Separable Residual Networks for Hyperspectral Image Classification

Zhao

Wang

et al. 2020

Complexity

View full text Add to dashboard Cite

show abstract

“…CNNs are a typical class of deep neural networks, [23, 24] and have strong feature-learning ability. A CNN can be divided into layers to extract features from a dataset.…”

Section: Methodsmentioning

confidence: 99%

Novel Spatiotemporal Feature Extraction Parallel Deep Neural Network for Forecasting Confirmed Cases of Coronavirus Disease 2019

Huang

Shen

Kuo

et al. 2020

Preprint

View full text Add to dashboard Cite

The coronavirus disease 2019 pandemic continues as of March 26 and spread to Europe on 10 approximately February 24. A report from April 29 revealed 1.26 million confirmed cases and 125 928 deaths 11 in Europe. This study proposed a novel deep neural network framework, COVID-19Net, which parallelly 12 combines a convolutional neural network (CNN) and bidirectional gated recurrent units (GRUs). Three 13 European countries with severe outbreaks were studied-Germany, Italy, and Spain-to extract 14 spatiotemporal feature and predict the number of confirmed cases. The prediction results acquired from 15 COVID-19Net were compared to those obtained using a CNN, GRU, and CNN-GRU. The mean absolute 16 error, mean absolute percentage error, and root mean square error, which are commonly used model 17 assessment indices, were used to compare the accuracy of the models. The results verified that COVID-19Net 18was notably more accurate than the other models. The mean absolute percentage error generated by COVID-19 19Net was 1.447 for Germany, 1.801 for Italy, and 2.828 for Spain, which were considerably lower than those 20 of the other models. This indicated that the proposed framework can accurately predict the accumulated 21 number of confirmed cases in the three countries and serve as a crucial reference for devising public health 22 strategies. 23Index Terms-Confirmed cases forecasting, COVID-19Net, parallel deep neural network 24 25

show abstract

“…Techniques to perform heartwood-sapwood distinction amount to extract features using PCA [2] or statistical properties of the wood texture [3] in both spatial and spectral dimensions, and then by classifying them using neural networks [4,2,3]. Recently, other approaches have exploited Deep Learning for HSI classification in the agriculture domain [5,6]. A typical challenge is the amount of training data and computational resources needed.…”

Section: Problem Statementmentioning

confidence: 99%

A Lightweight Approach For Wood Hyperspectral Images Classification

Htun

Boschetti²,

Buriro

et al. 2021

2021 IEEE International Conference on Multimedia &Amp; Expo Workshops (ICMEW)

View full text Add to dashboard Cite

This paper presents a Convolutional Neural Network (CNN)based spatial classifier to classify hyperspectral images for wood recognition. The spatial classifier is built by adapting the input and output units of Cifar10Net, a conventional image classifier that accepts three-band images as input. Obtained results in terms of accuracy and training time show that the proposed classifier can be trained using few training data, and few computational resources.

show abstract

Comparison of CNN Algorithms on Hyperspectral Image Classification in Agricultural Lands

Cited by 75 publications

References 43 publications

Hybrid Depth-Separable Residual Networks for Hyperspectral Image Classification

Hybrid Depth-Separable Residual Networks for Hyperspectral Image Classification

Novel Spatiotemporal Feature Extraction Parallel Deep Neural Network for Forecasting Confirmed Cases of Coronavirus Disease 2019

A Lightweight Approach For Wood Hyperspectral Images Classification

Contact Info

Product

Resources

About