scholarly journals A New Multispectral Data Augmentation Technique Based on Data Imputation

2021 ◽  
Vol 13 (23) ◽  
pp. 4875
Author(s):  
Álvaro Acción ◽  
Francisco Argüello ◽  
Dora B. Heras
Keyword(s):  
Remote Sensing ◽  
Data Augmentation ◽  
Matrix Completion ◽  
Computational Cost ◽  
High Energy ◽  
Classification Performance ◽  
Training Data ◽  
High Spectral Resolution ◽  
Data Imputation ◽  
Imputation Methods

Deep Learning (DL) has been recently introduced into the hyperspectral and multispectral image classification landscape. Despite the success of DL in the remote sensing field, DL models are computationally intensive due to the large number of parameters they need to learn. The high density of information present in remote sensing imagery with high spectral resolution can make the application of DL models to large scenes challenging. Methods such as patch-based classification require large amounts of data to be processed during the training and prediction stages, which translates into long processing times and high energy consumption. One of the solutions to decrease the computational cost of these models is to perform segment-based classification. Segment-based classification schemes can significantly decrease training and prediction times, and also offer advantages over simply reducing the size of the training datasets by randomly sampling training data. The lack of a large enough number of samples can, however, pose an additional challenge, causing these models to not generalize properly. Data augmentation methods are used to generate new synthetic samples based on existing data to increase the classification performance. In this work, we propose a new data augmentation scheme using data imputation and matrix completion methods for segment-based classification. The proposal has been validated using two high-resolution multispectral datasets from the literature. The results obtained show that the proposed approach successfully increases the classification performance across all the scenes tested and that data imputation methods applied to multispectral imagery are a valid means to perform data augmentation. A comparison of classification accuracy between different imputation methods applied to the proposed scheme was also carried out.

scholarly journals Using Vehicle Synthesis Generative Adversarial Networks to Improve Vehicle Detection in Remote Sensing Images

2019 ◽  
Vol 8 (9) ◽  
pp. 390 ◽  
Author(s):  
Kun Zheng ◽  
Mengfei Wei ◽  
Guangmin Sun ◽  
Bilal Anas ◽  
Yu Li
Keyword(s):  
Remote Sensing ◽  
Traffic Control ◽  
Data Augmentation ◽  
Vehicle Detection ◽  
Training Data ◽  
Generative Adversarial Networks ◽  
Remote Sensing Images ◽  
Adversarial Networks ◽  
Vehicle Data ◽  
Augmentation Strategy

Vehicle detection based on very high-resolution (VHR) remote sensing images is beneficial in many fields such as military surveillance, traffic control, and social/economic studies. However, intricate details about the vehicle and the surrounding background provided by VHR images require sophisticated analysis based on massive data samples, though the number of reliable labeled training data is limited. In practice, data augmentation is often leveraged to solve this conflict. The traditional data augmentation strategy uses a combination of rotation, scaling, and flipping transformations, etc., and has limited capabilities in capturing the essence of feature distribution and proving data diversity. In this study, we propose a learning method named Vehicle Synthesis Generative Adversarial Networks (VS-GANs) to generate annotated vehicles from remote sensing images. The proposed framework has one generator and two discriminators, which try to synthesize realistic vehicles and learn the background context simultaneously. The method can quickly generate high-quality annotated vehicle data samples and greatly helps in the training of vehicle detectors. Experimental results show that the proposed framework can synthesize vehicles and their background images with variations and different levels of details. Compared with traditional data augmentation methods, the proposed method significantly improves the generalization capability of vehicle detectors. Finally, the contribution of VS-GANs to vehicle detection in VHR remote sensing images was proved in experiments conducted on UCAS-AOD and NWPU VHR-10 datasets using up-to-date target detection frameworks.

scholarly journals PulseNetOne: Fast Unsupervised Pruning of Convolutional Neural Networks for Remote Sensing

2020 ◽  
Vol 12 (7) ◽  
pp. 1092
Author(s):  
David Browne ◽  
Michael Giering ◽  
Steven Prestwich
Keyword(s):  
Remote Sensing ◽  
Neural Networks ◽  
Deep Learning ◽  
Convolutional Neural Networks ◽  
Data Augmentation ◽  
Recognition Task ◽  
Scene Recognition ◽  
Training Data ◽  
Learning Approach ◽  
Scene Classification

Scene classification is an important aspect of image/video understanding and segmentation. However, remote-sensing scene classification is a challenging image recognition task, partly due to the limited training data, which causes deep-learning Convolutional Neural Networks (CNNs) to overfit. Another difficulty is that images often have very different scales and orientation (viewing angle). Yet another is that the resulting networks may be very large, again making them prone to overfitting and unsuitable for deployment on memory- and energy-limited devices. We propose an efficient deep-learning approach to tackle these problems. We use transfer learning to compensate for the lack of data, and data augmentation to tackle varying scale and orientation. To reduce network size, we use a novel unsupervised learning approach based on k-means clustering, applied to all parts of the network: most network reduction methods use computationally expensive supervised learning methods, and apply only to the convolutional or fully connected layers, but not both. In experiments, we set new standards in classification accuracy on four remote-sensing and two scene-recognition image datasets.

scholarly journals Multiple imputation of maritime search and rescue data at multiple missing patterns

PLoS ONE ◽  
2021 ◽  
Vol 16 (6) ◽  
pp. e0252129
Author(s):  
Guobo Wang ◽  
Minglu Ma ◽  
Lili Jiang ◽  
Fengyun Chen ◽  
Liansheng Xu
Keyword(s):  
Multiple Imputation ◽  
Expectation Maximization ◽  
Data Augmentation ◽  
Search And Rescue ◽  
Data Sets ◽  
Data Imputation ◽  
Imputation Methods ◽  
High Data ◽  
Data Missing ◽  
Maritime Search And Rescue

Based on the missing situation and actual needs of maritime search and rescue data, multiple imputation methods were used to construct complete data sets under different missing patterns. Probability density curves and overimputation diagnostics were used to explore the effects of multiple imputation. The results showed that the Data Augmentation (DA) algorithm had the characteristics of high operation efficiency and good imputation effect, but the algorithm was not suitable for data imputation when there was a high data missing rate. The EMB algorithm effectively restored the distribution of datasets with different data missing rates, and was less affected by the missing position; the EMB algorithm could obtain a good imputation effect even when there was a high data missing rate. Overimputation diagnostics could not only reflect the data imputation effect, but also show the correlation between different datasets, which was of great importance for deep data mining and imputation effect improvement. The Expectation-Maximization with Bootstrap (EMB) algorithm had a poor estimation effect on extreme data and failed to reflect the dataset’s variability characteristics.

scholarly journals Lidar Innovations for Technologies and Environmental Sciences (LITES) – An Remote Sensing Infrastructure Facility: Setup and Measurements Examples

2020 ◽  
Vol 237 ◽  
pp. 07017
Author(s):  
Boyan Tatarov ◽  
Detlef Müller ◽  
Matthias Tesche ◽  
Sung-Kyun Shin
Keyword(s):  
Remote Sensing ◽  
Spectral Resolution ◽  
High Energy ◽  
High Spectral Resolution ◽  
Ultra High Energy ◽  
Lidar Signal ◽  
Environmental Sciences ◽  
Time Resolved ◽  
The University

At the University of Hertfordshire, we have been developing a new remote sensing facility (LITES) to explore the feasibility of using Raman and/or fluorescence backscattering for chemical aerosol profiling. This paper provides an overview of the instruments of the facility and measurement examples. LITES includes a ultra-high-energy Nd:YAG/OPO setup, spectroscopic equipment with high spectral resolution, several imaging and single detectors that allow for time-resolved (lidar) signal detection, a Raman/fluorescence microscope, and a suite of gas and aerosol chambers. We present examples of elastic, rotational and vibrational spectroscopic lidar signals, as well as in-situ microscopic spectrums of dust and bio-aerosol compounds.

scholarly journals Vision Transformers for Remote Sensing Image Classification

2021 ◽  
Vol 13 (3) ◽  
pp. 516
Author(s):  
Yakoub Bazi ◽  
Laila Bashmal ◽  
Mohamad M. Al Rahhal ◽  
Reham Al Dayil ◽  
Naif Al Ajlan
Keyword(s):  
Remote Sensing ◽  
Language Processing ◽  
Additional Data ◽  
Data Augmentation ◽  
State Of The Art ◽  
Remote Sensing Image ◽  
Classification Performance ◽  
Scene Classification ◽  
Remote Sensing Image Classification ◽  
Augmentation Strategies

In this paper, we propose a remote-sensing scene-classification method based on vision transformers. These types of networks, which are now recognized as state-of-the-art models in natural language processing, do not rely on convolution layers as in standard convolutional neural networks (CNNs). Instead, they use multihead attention mechanisms as the main building block to derive long-range contextual relation between pixels in images. In a first step, the images under analysis are divided into patches, then converted to sequence by flattening and embedding. To keep information about the position, embedding position is added to these patches. Then, the resulting sequence is fed to several multihead attention layers for generating the final representation. At the classification stage, the first token sequence is fed to a softmax classification layer. To boost the classification performance, we explore several data augmentation strategies to generate additional data for training. Moreover, we show experimentally that we can compress the network by pruning half of the layers while keeping competing classification accuracies. Experimental results conducted on different remote-sensing image datasets demonstrate the promising capability of the model compared to state-of-the-art methods. Specifically, Vision Transformer obtains an average classification accuracy of 98.49%, 95.86%, 95.56% and 93.83% on Merced, AID, Optimal31 and NWPU datasets, respectively. While the compressed version obtained by removing half of the multihead attention layers yields 97.90%, 94.27%, 95.30% and 93.05%, respectively.

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.

scholarly journals Improving Object Tracking by Added Noise and Channel Attention

Sensors ◽  
2020 ◽  
Vol 20 (13) ◽  
pp. 3780 ◽  
Author(s):  
Mustansar Fiaz ◽  
Arif Mahmood ◽  
Ki Yeol Baek ◽  
Sehar Shahzad Farooq ◽  
Soon Ki Jung
Keyword(s):  
Large Scale ◽  
Data Augmentation ◽  
Feature Fusion ◽  
State Of The Art ◽  
Computational Cost ◽  
Training Data ◽  
Superior Performance ◽  
Input Noise ◽  
Offline Learning ◽  
Benchmark Datasets

CNN-based trackers, especially those based on Siamese networks, have recently attracted considerable attention because of their relatively good performance and low computational cost. For many Siamese trackers, learning a generic object model from a large-scale dataset is still a challenging task. In the current study, we introduce input noise as regularization in the training data to improve generalization of the learned model. We propose an Input-Regularized Channel Attentional Siamese (IRCA-Siam) tracker which exhibits improved generalization compared to the current state-of-the-art trackers. In particular, we exploit offline learning by introducing additive noise for input data augmentation to mitigate the overfitting problem. We propose feature fusion from noisy and clean input channels which improves the target localization. Channel attention integrated with our framework helps finding more useful target features resulting in further performance improvement. Our proposed IRCA-Siam enhances the discrimination of the tracker/background and improves fault tolerance and generalization. An extensive experimental evaluation on six benchmark datasets including OTB2013, OTB2015, TC128, UAV123, VOT2016 and VOT2017 demonstrate superior performance of the proposed IRCA-Siam tracker compared to the 30 existing state-of-the-art trackers.

scholarly journals Effect of De-noising by Wavelet Filtering and Data Augmentation by Borderline SMOTE on the Classification of Imbalanced Datasets of Pig Behavior

2021 ◽  
Vol 2 ◽  
Author(s):  
Min Jin ◽  
Chunguang Wang ◽  
Dan Børge Jensen
Keyword(s):  
Data Augmentation ◽  
Science Research ◽  
Classification Performance ◽  
Training Data ◽  
Misclassification Rate ◽  
Feed Forward Neural Network ◽  
Animal Science ◽  
Imbalanced Datasets ◽  
Wavelet Filtering

Classification of imbalanced datasets of animal behavior has been one of the top challenges in the field of animal science. An imbalanced dataset will lead many classification algorithms to being less effective and result in a higher misclassification rate for the minority classes. The aim of this study was to assess a method for addressing the problem of imbalanced datasets of pigs' behavior by using an over-sampling method, namely Borderline-SMOTE. The pigs' activity was measured using a triaxial accelerometer, which was mounted on the back of the pigs. Wavelet filtering and Borderline-SMOTE were both applied as methods to pre-process the dataset. A multilayer feed-forward neural network was trained and validated with 21 input features to classify four pig activities: lying, standing, walking, and exploring. The results showed that wavelet filtering and Borderline-SMOTE both lead to improved performance. Furthermore, Borderline-SMOTE yielded greater improvements in classification performance than an alternative method for balancing the training data, namely random under-sampling, which is commonly used in animal science research. However, the overall performance was not adequate to satisfy the research needs in this field and to address the common but urgent problem of imbalanced behavior dataset.

scholarly journals Data Augmentation: Using Channel-Level Recombination to Improve Classification Performance for Motor Imagery EEG

2021 ◽  
Vol 15 ◽  
Author(s):  
Yu Pei ◽  
Zhiguo Luo ◽  
Ye Yan ◽  
Huijiong Yan ◽  
Jing Jiang ◽  
...  
Keyword(s):  
Deep Learning ◽  
Motor Imagery ◽  
Data Augmentation ◽  
Brain Area ◽  
Comparative Investigation ◽  
Classification Performance ◽  
Training Data ◽  
Learning Technology ◽  
Data Generation ◽  
Public Datasets

The quality and quantity of training data are crucial to the performance of a deep-learning-based brain-computer interface (BCI) system. However, it is not practical to record EEG data over several long calibration sessions. A promising time- and cost-efficient solution is artificial data generation or data augmentation (DA). Here, we proposed a DA method for the motor imagery (MI) EEG signal called brain-area-recombination (BAR). For the BAR, each sample was first separated into two ones (named half-sample) by left/right brain channels, and the artificial samples were generated by recombining the half-samples. We then designed two schemas (intra- and adaptive-subject schema) corresponding to the single- and multi-subject scenarios. Extensive experiments using the classifier of EEGnet were conducted on two public datasets under various training set sizes. In both schemas, the BAR method can make the EEGnet have a better performance of classification (p < 0.01). To make a comparative investigation, we selected two common DA methods (noise-added and flipping), and the BAR method beat them (p < 0.05). Further, using the proposed BAR for augmentation, EEGnet achieved up to 8.3% improvement than a typical decoding algorithm CSP-SVM (p < 0.01), note that both the models were trained on the augmented dataset. This study shows that BAR usage can significantly improve the classification ability of deep learning to MI-EEG signals. To a certain extent, it may promote the development of deep learning technology in the field of BCI.

scholarly journals Use of Hyper-Spectral Visible and Near-Infrared Satellite Data for Timely Estimates of the Earth’s Surface Reflectance in Cloudy and Aerosol Loaded Conditions: Part 1–Application to RGB Image Restoration Over Land With GOME-2

2022 ◽  
Vol 2 ◽  
Author(s):  
J. Joiner ◽  
Z. Fasnacht ◽  
W. Qin ◽  
Y. Yoshida ◽  
A. P. Vasilkov ◽  
...  
Keyword(s):  
Remote Sensing ◽  
Precision Agriculture ◽  
Land Surface ◽  
Rayleigh Scattering ◽  
Near Infrared ◽  
Computational Cost ◽  
Training Data ◽  
Optical Remote Sensing ◽  
Spectral Features ◽  
Hyper Spectral

Space-based quantitative passive optical remote sensing of the Earth’s surface typically involves the detection and elimination of cloud-contaminated pixels as an initial processing step. We explore a fundamentally different approach; we use machine learning with cloud contaminated satellite hyper-spectral data to estimate underlying terrestrial surface reflectances at red, green, and blue (RGB) wavelengths. An artificial neural network (NN) reproduces land RGB reflectances with high fidelity, even in scenes with moderate to high cloud optical thicknesses. This implies that spectral features of the Earth’s surface can be detected and distinguished in the presence of clouds, even when they are partially and visibly obscured by clouds; the NN is able to separate the spectral fingerprint of the Earth’s surface from that of the clouds, aerosols, gaseous absorption, and Rayleigh scattering, provided that there are adequately different spectral features and that the clouds are not completely opaque. Once trained, the NN enables rapid estimates of RGB reflectances with little computational cost. Aside from the training data, there is no requirement of prior information regarding the land surface spectral reflectance, nor is there need for radiative transfer calculations. We test different wavelength windows and instrument configurations for reconstruction of surface reflectances. This work provides an initial example of a general approach that has many potential applications in land and ocean remote sensing as well as other practical uses such as in search and rescue, precision agriculture, and change detection.

Sign in / Sign up

Export Citation Format

Share Document