scholarly journals Data Augmentation and Spectral Structure Features for Limited Samples Hyperspectral Classification

2021 ◽  
Vol 13 (4) ◽  
pp. 547
Author(s):  
Wenning Wang ◽  
Xuebin Liu ◽  
Xuanqin Mou
Keyword(s):  
Classification Accuracy ◽  
Data Augmentation ◽  
Classification Problem ◽  
Classification Performance ◽  
Spectral Structure ◽  
Limited Sample ◽  
Sample Classification ◽  
Training Samples ◽  
Traditional Classification ◽  
Hyperspectral Classification

For both traditional classification and current popular deep learning methods, the limited sample classification problem is very challenging, and the lack of samples is an important factor affecting the classification performance. Our work includes two aspects. First, the unsupervised data augmentation for all hyperspectral samples not only improves the classification accuracy greatly with the newly added training samples, but also further improves the classification accuracy of the classifier by optimizing the augmented test samples. Second, an effective spectral structure extraction method is designed, and the effective spectral structure features have a better classification accuracy than the true spectral features.

scholarly journals Effects of Class Purity of Training Patch on Classification Performance of Crop Classification with Convolutional Neural Network

2020 ◽  
Vol 10 (11) ◽  
pp. 3773
Author(s):  
Soyeon Park ◽  
No-Wook Park
Keyword(s):  
Classification Accuracy ◽  
Supervised Classification ◽  
Patch Size ◽  
Classification Performance ◽  
Spatial Homogeneity ◽  
High Class ◽  
Large Size ◽  
Training Samples ◽  
Compositional Homogeneity ◽  
Crop Classification

As the performance of supervised classification using convolutional neural networks (CNNs) are affected significantly by training patches, it is necessary to analyze the effects of the information content of training patches in patch-based classification. The objective of this study is to quantitatively investigate the effects of class purity of a training patch on performance of crop classification. Here, class purity that refers to a degree of compositional homogeneity of classes within a training patch is considered as a primary factor for the quantification of information conveyed by training patches. New quantitative indices for class homogeneity and variations of local class homogeneity over the study area are presented to characterize the spatial homogeneity of the study area. Crop classification using 2D-CNN was conducted in two regions (Anbandegi in Korea and Illinois in United States) with distinctive spatial distributions of crops and class homogeneity over the area to highlight the effect of class purity of a training patch. In the Anbandegi region with high class homogeneity, superior classification accuracy was obtained when using large size training patches with high class purity (7.1%p improvement in overall accuracy over classification with the smallest patch size and the lowest class purity). Training patches with high class purity could yield a better identification of homogenous crop parcels. In contrast, using small size training patches with low class purity yielded the highest classification accuracy in the Illinois region with low class homogeneity (19.8%p improvement in overall accuracy over classification with the largest patch size and the highest class purity). Training patches with low class purity could provide useful information for the identification of diverse crop parcels. The results indicate that training samples in patch-based classification should be selected based on the class purity that reflects the local class homogeneity of the study area.

scholarly journals Multiscale Information Fusion for Hyperspectral Image Classification Based on Hybrid 2D-3D CNN

2021 ◽  
Vol 13 (12) ◽  
pp. 2268
Author(s):  
Hang Gong ◽  
Qiuxia Li ◽  
Chunlai Li ◽  
Haishan Dai ◽  
Zhiping He ◽  
...  
Keyword(s):  
Classification Accuracy ◽  
Spatial Information ◽  
Hyperspectral Image ◽  
Small Sample ◽  
Hyperspectral Images ◽  
Sample Problem ◽  
Training Samples ◽  
Small Sample Problem ◽  
3D Cnn ◽  
Hyperspectral Classification

Hyperspectral images are widely used for classification due to its rich spectral information along with spatial information. To process the high dimensionality and high nonlinearity of hyperspectral images, deep learning methods based on convolutional neural network (CNN) are widely used in hyperspectral classification applications. However, most CNN structures are stacked vertically in addition to using a onefold size of convolutional kernels or pooling layers, which cannot fully mine the multiscale information on the hyperspectral images. When such networks meet the practical challenge of a limited labeled hyperspectral image dataset—i.e., “small sample problem”—the classification accuracy and generalization ability would be limited. In this paper, to tackle the small sample problem, we apply the semantic segmentation function to the pixel-level hyperspectral classification due to their comparability. A lightweight, multiscale squeeze-and-excitation pyramid pooling network (MSPN) is proposed. It consists of a multiscale 3D CNN module, a squeezing and excitation module, and a pyramid pooling module with 2D CNN. Such a hybrid 2D-3D-CNN MSPN framework can learn and fuse deeper hierarchical spatial–spectral features with fewer training samples. The proposed MSPN was tested on three publicly available hyperspectral classification datasets: Indian Pine, Salinas, and Pavia University. Using 5%, 0.5%, and 0.5% training samples of the three datasets, the classification accuracies of the MSPN were 96.09%, 97%, and 96.56%, respectively. In addition, we also selected the latest dataset with higher spatial resolution, named WHU-Hi-LongKou, as the challenge object. Using only 0.1% of the training samples, we could achieve a 97.31% classification accuracy, which is far superior to the state-of-the-art hyperspectral classification methods.

scholarly journals Spatial-Adaptive Siamese Residual Network for Multi-/Hyperspectral Classification

2020 ◽  
Vol 12 (10) ◽  
pp. 1640 ◽  
Author(s):  
Zhi He ◽  
Dan He
Keyword(s):  
Large Scale ◽  
Nearest Neighbor ◽  
Three Dimensional ◽  
Classification Performance ◽  
Majority Voting ◽  
Residual Network ◽  
Training Samples ◽  
Simple Linear Iterative Clustering ◽  
Limited Training Samples ◽  
Hyperspectral Classification

Deep learning methods have been successfully applied for multispectral and hyperspectral images classification due to their ability to extract hierarchical abstract features. However, the performance of these methods relies heavily on large-scale training samples. In this paper, we propose a three-dimensional spatial-adaptive Siamese residual network (3D-SaSiResNet) that requires fewer samples and still enhances the performance. The proposed method consists of two main steps: construction of 3D spatial-adaptive patches and Siamese residual network for multiband images classification. In the first step, the spectral dimension of the original multiband images is reduced by a stacked autoencoder and superpixels of each band are obtained by the simple linear iterative clustering (SLIC) method. Superpixels of the original multiband image can be finally generated by majority voting. Subsequently, the 3D spatial-adaptive patch of each pixel is extracted from the original multiband image by reference to the previously generated superpixels. In the second step, a Siamese network composed of two 3D residual networks is designed to extract discriminative features for classification and we train the 3D-SaSiResNet by pairwise inputting the training samples into the networks. The testing samples are then fed into the trained 3D-SaSiResNet and the learned features of the testing samples are classified by the nearest neighbor classifier. Experimental results on three multiband image datasets show the feasibility of the proposed method in enhancing classification performance even with limited training samples.

scholarly journals A Convolutional Neural Network with Fletcher–Reeves Algorithm for Hyperspectral Image Classification

2019 ◽  
Vol 11 (11) ◽  
pp. 1325 ◽  
Author(s):  
Chen Chen ◽  
Yi Ma ◽  
Guangbo Ren
Keyword(s):  
Image Classification ◽  
Classification Accuracy ◽  
Hyperspectral Image ◽  
Training Sample ◽  
Practical Applications ◽  
Model Based ◽  
Training Samples ◽  
Proposed Model ◽  
Convergence Performance ◽  
Hyperspectral Classification

Deep learning models, especially the convolutional neural networks (CNNs), are very active in hyperspectral remote sensing image classification. In order to better apply the CNN model to hyperspectral classification, we propose a CNN model based on Fletcher–Reeves algorithm (F–R CNN), which uses the Fletcher–Reeves (F–R) algorithm for gradient updating to optimize the convergence performance of the model in classification. In view of the fact that there are fewer optional training samples in practical applications, we further propose a method of increasing the number of samples by adding a certain degree of perturbed samples, which can also test the anti-interference ability of classification methods. Furthermore, we analyze the anti-interference and convergence performance of the proposed model in terms of different training sample data sets, different batch training sample numbers and iteration time. In this paper, we describe the experimental process in detail and comprehensively evaluate the proposed model based on the classification of CHRIS hyperspectral imagery covering coastal wetlands, and further evaluate it on a commonly used hyperspectral image benchmark dataset. The experimental results show that the accuracy of the two models after increasing training samples and adjusting the number of batch training samples is improved. When the number of batch training samples is continuously increased to 350, the classification accuracy of the proposed method can still be maintained above 80.7%, which is 2.9% higher than the traditional one. And its time consumption is less than that of the traditional one while ensuring classification accuracy. It can be concluded that the proposed method has anti-interference ability and outperforms the traditional CNN in terms of batch computing adaptability and convergence speed.

scholarly journals Study on the Classification Performance of Underwater Sonar Image Classification Based on Convolutional Neural Networks for Detecting a Submerged Human Body

Sensors ◽  
2019 ◽  
Vol 20 (1) ◽  
pp. 94 ◽  
Author(s):  
Huu-Thu Nguyen ◽  
Eon-Ho Lee ◽  
Sejin Lee
Keyword(s):  
Human Body ◽  
Classification Accuracy ◽  
Data Augmentation ◽  
Incidence Angle ◽  
Light Condition ◽  
Classification Performance ◽  
Training Data ◽  
Sonar Image ◽  
Sonar Sensor ◽  
Sonar Images

Auto-detecting a submerged human body underwater is very challenging with the absolute necessity to a diver or a submersible. For the vision sensor, the water turbidity and limited light condition make it difficult to take clear images. For this reason, sonar sensors are mainly utilized in water. However, even though a sonar sensor can give a plausible underwater image within this limitation, the sonar image’s quality varies greatly depending on the background of the target. The readability of the sonar image is very different according to the target distance from the underwater floor or the incidence angle of the sonar sensor to the floor. The target background must be very considerable because it causes scattered and polarization noise in the sonar image. To successfully classify the sonar image with these noises, we adopted a Convolutional Neural Network (CNN) such as AlexNet and GoogleNet. In preparing the training data for this model, the data augmentation on scattering and polarization were implemented to improve the classification accuracy from the original sonar image. It could be practical to classify sonar images undersea even by training sonar images only from the simple testbed experiments. Experimental validation was performed using three different datasets of underwater sonar images from a submerged body of a dummy, resulting in a final average classification accuracy of 91.6% using GoogleNet.

Tire Defects Classification with Multi-Contrast Convolutional Neural Networks

2017 ◽  
Vol 32 (04) ◽  
pp. 1850011 ◽  
Author(s):  
Xuehong Cui ◽  
Yun Liu ◽  
Yan Zhang ◽  
Chuanxu Wang
Keyword(s):  
Neural Networks ◽  
Classification Accuracy ◽  
Data Augmentation ◽  
State Of The Art ◽  
High Accuracy ◽  
Defect Classification ◽  
Average Accuracy ◽  
Training Samples ◽  
Limited Training Samples ◽  
Defects Classification

The objective of this study is to improve the accuracy in tire defect classification with limited training samples under varying illuminations. We investigate an algorithm based on deep learning to achieve high accuracy with limited samples. First, image contrast normalizations and data augmentation were used to avoid overfitting problems of the network with a large number of parameters. Furthermore, multi-column CNN is proposed by combining several CNNs trained on differently preprocessed data into a multi-column CNN (MC-CNN), and then their predictions are averaged as the output of the proposed network. An average accuracy of 98.47% is achieved with the proposed CNN-based method. Experimental results show that our scheme receives satisfactory classification accuracy and outperforms state-of-the-art methods on the same tire defect dataset.

scholarly journals Hybrid Dense Network with Dual Attention for Hyperspectral Image Classification

2021 ◽  
Vol 13 (23) ◽  
pp. 4921
Author(s):  
Jinling Zhao ◽  
Lei Hu ◽  
Yingying Dong ◽  
Linsheng Huang
Keyword(s):  
Classification Accuracy ◽  
Hyperspectral Image ◽  
Network Models ◽  
Classification Performance ◽  
Dense Network ◽  
Feature Maps ◽  
Neural Network Models ◽  
Practical Applications ◽  
Training Samples ◽  
3D Cnn

Hyperspectral images (HSIs) have been widely used in many fields of application, but it is still extremely challenging to obtain higher classification accuracy, especially when facing a smaller number of training samples in practical applications. It is very time-consuming and laborious to acquire enough labeled samples. Consequently, an efficient hybrid dense network was proposed based on a dual-attention mechanism, due to limited training samples and unsatisfactory classification accuracy. The stacked autoencoder was first used to reduce the dimensions of HSIs. A hybrid dense network framework with two feature-extraction branches was then established in order to extract abundant spectral–spatial features from HSIs, based on the 3D and 2D convolutional neural network models. In addition, spatial attention and channel attention were jointly introduced in order to achieve selective learning of features derived from HSIs. The feature maps were further refined, and more important features could be retained. To improve computational efficiency and prevent the overfitting, the batch normalization layer and the dropout layer were adopted. The Indian Pines, Pavia University, and Salinas datasets were selected to evaluate the classification performance; 5%, 1%, and 1% of classes were randomly selected as training samples, respectively. In comparison with the REF-SVM, 3D-CNN, HybridSN, SSRN, and R-HybridSN, the overall accuracy of our proposed method could still reach 96.80%, 98.28%, and 98.85%, respectively. Our results show that this method can achieve a satisfactory classification performance even in the case of fewer training samples.

scholarly journals Remote Sensing Image Scene Classification via Label Augmentation and Intra-Class Constraint

2021 ◽  
Vol 13 (13) ◽  
pp. 2566
Author(s):  
Hao Xie ◽  
Yushi Chen ◽  
Pedram Ghamisi
Keyword(s):  
Remote Sensing ◽  
Classification Accuracy ◽  
Data Augmentation ◽  
Original Data ◽  
Small Data ◽  
Scene Classification ◽  
Training Set ◽  
Output Distribution ◽  
Training Samples ◽  
Leibler Divergence

In recent years, many convolutional neural network (CNN)-based methods have been proposed to address the scene classification tasks of remote sensing images. Since the number of training samples in RS datasets is generally small, data augmentation is often used to expand the training set. It is, however, not appropriate when original data augmentation methods keep the label and change the content of the image at the same time. In this study, label augmentation (LA) is presented to fully utilize the training set by assigning a joint label to each generated image, which considers the label and data augmentation at the same time. Moreover, the output of images obtained by different data augmentation is aggregated in the test process. However, the augmented samples increase the intra-class diversity of the training set, which is a challenge to complete the following classification process. To address the above issue and further improve classification accuracy, Kullback–Leibler divergence (KL) is used to constrain the output distribution of two training samples with the same scene category to generate a consistent output distribution. Extensive experiments were conducted on widely-used UCM, AID and NWPU datasets. The proposed method can surpass the other state-of-the-art methods in terms of classification accuracy. For example, on the challenging NWPU dataset, competitive overall accuracy (i.e., 91.05%) is obtained with a 10% training ratio.

scholarly journals Fuzzy ARTMAP Ensemble Based Decision Making and Application

2013 ◽  
Vol 2013 ◽  
pp. 1-7
Author(s):  
Min Jin ◽  
Zengbing Xu ◽  
Ren Li ◽  
Dan Wu
Keyword(s):  
Decision Making ◽  
Classification Accuracy ◽  
Classification Performance ◽  
Fuzzy Artmap ◽  
Final Decision ◽  
Ensemble Approach ◽  
Training Samples ◽  
Sample Presentation

Because the performance of single FAM is affected by the sequence of sample presentation for the offline mode of training, a fuzzy ARTMAP (FAM) ensemble approach based on the improved Bayesian belief method is supposed to improve the classification accuracy. The training samples are input into a committee of FAMs in different sequence, the output from these FAMs is combined, and the final decision is derived by the improved Bayesian belief method. The experiment results show that the proposed FAMs’ ensemble can classify the different category reliably and has a better classification performance compared with single FAM.

ER rule classifier with an optimization operator recommendation

2021 ◽  
pp. 1-13
Author(s):  
Xiaoyan Wang ◽  
Jianbin Sun ◽  
Qingsong Zhao ◽  
Yaqian You ◽  
Jiang Jiang
Keyword(s):  
Classification Accuracy ◽  
Comprehensive Analysis ◽  
Small Sample ◽  
Empirical Knowledge ◽  
Classification Methods ◽  
Turbofan Engine ◽  
Training Samples ◽  
Optimization Operator ◽  
Recommendation Strategy

It is difficult for many classic classification methods to consider expert experience and classify small-sample datasets well. The evidential reasoning rule (ER rule) classifier can solve these problems. The ER rule has strong processing and comprehensive analysis abilities for diversified mixed information and can solve problems with expert experience effectively. Moreover, the initial parameters of the classifier constructed based on the ER rule can be set according to empirical knowledge instead of being trained by a large number of samples, which can help the classifier classify small-sample datasets well. However, the initial parameters of the ER rule classifier need to be optimized, and choosing the best optimization algorithm is still a challenge. Considering these problems, the ER rule classifier with an optimization operator recommendation is proposed in this paper. First, the initial ER rule classifier is constructed based on training samples and expert experience. Second, the adjustable parameters are optimized, in which the optimization operator recommendation strategy is applied to select the best algorithm by partial samples, and then experiments with full samples are carried out. Finally, a case study on a turbofan engine degradation simulation dataset is carried out, and the results indicate that the ER rule classifier has a higher classification accuracy than other classic classifiers, which demonstrates the capability and effectiveness of the proposed ER rule classifier with an optimization operator recommendation.

Sign in / Sign up

Export Citation Format

Share Document