Self-Taught Feature Learning for Hyperspectral Image Classification

Kemker, Ronald; Kanan, Christopher

doi:10.1109/tgrs.2017.2651639

Cited by 110 publications

(92 citation statements)

References 49 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In recent years, transfer learning has been successfully applied in many fields, especially in the remote sensing field [114], [115] where the available training samples are not sufficient. In [77], [104], [105], [107], [108], transfer learning was employed to set the initial values of network parameters copied from other trained deep networks, which provides better classification performance compared with the random initialization-based methods. Fig.…”

Section: B Transfer Learningmentioning

confidence: 99%

“…For example, research works [35], [37], [53], [54], [73], [104], [111], [116] utilized a fully connected network to classify HSIs, where the training of the network was divided into pre-training in an unsupervised manner and fine-tuning with labeled data. In the pre-training stage, the sample from unlabeled data set was first mapped into an intermediate feature.…”

Section: Unsupervised/semi-supervised Feature Learningmentioning

confidence: 99%

See 1 more Smart Citation

Deep Learning for Hyperspectral Image Classification: An Overview

Li¹,

Song²,

Fang³

et al. 2019

IEEE Trans. Geosci. Remote Sensing

1,331

452

View full text Add to dashboard Cite

Hyperspectral image (HSI) classification has become a hot topic in the field of remote sensing. In general, the complex characteristics of hyperspectral data make the accurate classification of such data challenging for traditional machine learning methods. In addition, hyperspectral imaging often deals with an inherently nonlinear relation between the captured spectral information and the corresponding materials. In recent years, deep learning has been recognized as a powerful feature-extraction tool to effectively address nonlinear problems and widely used in a number of image processing tasks. Motivated by those successful applications, deep learning has also been introduced to classify HSIs and demonstrated good performance. This survey paper presents a systematic review of deep learning-based HSI classification literatures and compares several strategies for this topic. Specifically, we first summarize the main challenges of HSI classification which cannot be effectively overcome by traditional machine learning methods, and also introduce the advantages of deep learning to handle these problems. Then, we build a framework which divides the corresponding works into spectralfeature networks, spatial-feature networks, and spectral-spatialfeature networks to systematically review the recent achievements in deep learning-based HSI classification. In addition, considering the fact that available training samples in the remote sensing field are usually very limited and training deep networks require a large number of samples, we include some strategies to improve classification performance, which can provide some guidelines for future studies on this topic. Finally, several representative deep learning-based classification methods are conducted on real HSIs in our experiments.

show abstract

Section: B Transfer Learningmentioning

confidence: 99%

Section: Unsupervised/semi-supervised Feature Learningmentioning

confidence: 99%

Deep Learning for Hyperspectral Image Classification: An Overview

Li¹,

Song²,

Fang³

et al. 2019

IEEE Trans. Geosci. Remote Sensing

1,331

452

View full text Add to dashboard Cite

show abstract

“…There has been interest in using CNNs for analyzing remote sensing imagery [3], [4], [5], [6], [7]. Uzkent et al adapted correlation filters trained on RGB images with HSI bands to successfully track cars in moving platform scenarios [3].…”

Section: Introductionmentioning

confidence: 99%

“…Kleynhans et al compared the performance of predicting top-of-atmosphere thermal radiance by using forward modeling with radiosonde data and radiative transfer modeling (MODTRAN [8]) against a multi-layer perception (MLP) and CNN and observed better performance from MLP and CNN in all experimental cases [5]. Kemker et al used multi-scale independent component analysis and stacked convolutional autoencoders as self-supervised learning tasks before performing semantic segmentation on multispectral and hyperspectral imagery [6], [7].…”

Section: Introductionmentioning

confidence: 99%

AeroRIT: A New Scene for Hyperspectral Image Analysis

Rangnekar

Mokashi

Ientilucci

et al. 2020

IEEE Trans. Geosci. Remote Sensing

Self Cite

View full text Add to dashboard Cite

We investigate modifying convolutional neural network (CNN) architecture to facilitate aerial hyperspectral scene understanding and present a new hyperspectral dataset-AeroRITthat is large enough for CNN training. To date the majority of hyperspectral airborne have been confined to various subcategories of vegetation and roads and this scene introduces two new categories: buildings and cars. To the best of our knowledge, this is the first comprehensive large-scale hyperspectral scene with nearly seven million pixel annotations for identifying cars, roads, and buildings. We compare the performance of three popular architectures -SegNet, U-Net, and Res-U-Net, for scene understanding and object identification via the task of dense semantic segmentation to establish a benchmark for the scene. To further strengthen the network, we add squeeze and excitation blocks for better channel interactions and use self-supervised learning for better encoder initialization. Aerial hyperspectral image analysis has been restricted to small datasets with limited train/test splits capabilities and we believe that AeroRIT will help advance the research in the field with a more complex object distribution to perform well on. The full dataset, with flight lines in radiance and reflectance domain, is available for download at https://github.com/aneesh3108/AeroRIT. This dataset is the first step towards developing robust algorithms for hyperspectral airborne sensing that can robustly perform advanced tasks like vehicle tracking and occlusion handling.

show abstract

“…Unsupervised models (e.g., auto-encoders), are trained to extract features from large unlabeled data sets. By restricting the encoder-decoder structure, one can adjust the model to achieve the results of the required task [6] [7] [8]. Supervised models (e.g., convolutional neural networks (CNNs) and deep belief networks) are trained using ground truth (GT) information as expected output of the network.…”

Section: Introductionmentioning

confidence: 99%

Ground Truth Simulation for Deep Learning Classification of Mid-Resolution Venus Images Via Unmixing of High-Resolution Hyperspectral Fenix Data

Faran

Netanyahu

David

et al. 2019

IGARSS 2019 - 2019 IEEE International Geoscience and Remote Sensing Symposium

View full text Add to dashboard Cite

Training a deep neural network for classification constitutes a major problem in remote sensing due to the lack of adequate field data. Acquiring high-resolution ground truth (GT) by human interpretation is both cost-ineffective and inconsistent. We propose, instead, to utilize high-resolution, hyperspectral images for solving this problem, by unmixing these images to obtain reliable GT for training a deep network. Specifically, we simulate GT from high-resolution, hyperspectral FENIX images, and use it for training a convolutional neural network (CNN) for pixel-based classification. We show how the model can be transferred successfully to classify new mid-resolution VENµS imagery.

show abstract

Self-Taught Feature Learning for Hyperspectral Image Classification

Cited by 110 publications

References 49 publications

Deep Learning for Hyperspectral Image Classification: An Overview

Deep Learning for Hyperspectral Image Classification: An Overview

AeroRIT: A New Scene for Hyperspectral Image Analysis

Ground Truth Simulation for Deep Learning Classification of Mid-Resolution Venus Images Via Unmixing of High-Resolution Hyperspectral Fenix Data

Contact Info

Product

Resources

About