Advances in computing technology have fostered the development of new and powerful deep learning (DL) techniques, which have demonstrated promising results in a wide range of applications. Particularly, DL methods have been successfully used to classify remotely sensed data collected by Earth Observation (EO) instruments. Hyperspectral imaging (HSI) is a hot topic in remote sensing data analysis due to the vast amount of information comprised by this kind of images, which allows for a better characterization and exploitation of the Earth surface by combining rich spectral and spatial information. However, HSI poses major challenges for supervised classification methods due to the high dimensionality of the data and the limited availability of training samples. These issues, together with the high intraclass variability (and interclass similarity) –often present in HSI data– may hamper the effectiveness of classifiers. In order to solve these limitations, several DL-based architectures have been recently developed, exhibiting great potential in HSI data interpretation. This paper provides a comprehensive review of the current-state-of-the-art in DL for HSI classification, analyzing the strengths and weaknesses of the most widely used classifiers in the literature. For each discussed method, we provide quantitative results using several well-known and widely used HSI scenes, thus providing an exhaustive comparison of the discussed techniques. The paper concludes with some remarks and hints about future challenges in the application of DL techniques to HSI classification. The source codes of the methods discussed in this paper are available from: https://github.com/mhaut/hyperspectral_deeplearning_review .

Deep learning classifiers for hyperspectral imaging: A review

Semantic segmentation of remote sensing imagery has been employed in many applications and is a key research topic for decades. With the success of deep learning methods in the field of computer vision, researchers have made a great effort to transfer their superior performance to the field of remote sensing image analysis. This paper starts with a summary of the fundamental deep neural network architectures and reviews the most recent developments of deep learning methods for semantic segmentation of remote sensing imagery including non-conventional data such as hyperspectral images and point clouds. In our review of the literature, we identified three major challenges faced by researchers and summarize the innovative development to address them. As tremendous efforts have been devoted to advancing pixel-level accuracy, the emerged deep learning methods demonstrated much-improved performance on several public data sets. As to handling the non-conventional, unstructured point cloud and rich spectral imagery, the performance of the state-of-the-art methods is, on average, inferior to that of the satellite imagery. Such a performance gap also exists in learning from small data sets. In particular, the limited non-conventional remote sensing data sets with labels is an obstacle to developing and evaluating new deep learning methods.

A review of deep learning methods for semantic segmentation of remote sensing imagery

With the advent of big data era, deep learning (DL) has become an essential research subject in the field of artificial intelligence (AI). DL algorithms are characterized with powerful feature learning and expression capabilities compared with the traditional machine learning (ML) methods, which attracts worldwide researchers from different fields to its increasingly wide applications. Furthermore, in the field of geochnical engineering, DL has been widely adopted in various research topics, a comprehensive review summarizing its application is desirable. Consequently, this study presented the state of practice of DL in geotechnical engineering, and depicted the statistical trend of the published papers. Four major algorithms, including feedforward neural (FNN), recurrent neural network (RNN), convolutional neural network (CNN) and generative adversarial network (GAN) along with their geotechnical applications were elaborated. In addition, a thorough summary containing pubilished literatures, the corresponding reference cases, the adopted DL algorithms as well as the related geotechnical topics was compiled. Furthermore, the challenges and perspectives of future development of DL in geotechnical engineering were presented and discussed.

Application of deep learning algorithms in geotechnical engineering: a short critical review

Deep support vector machine for hyperspectral image classification

The rapid development of deep learning brings effective solutions for remote sensing image interpretation. Training deep neural network models usually require a large number of manually labeled samples. However, there is a limitation to obtain sufficient labeled samples in remote sensing field to satisfy the data requirement. Therefore, it is of great significance to conduct the research on few-shot learning for remote sensing image interpretation. First, this article provides a bibliometric analysis of the existing works for remote sensing interpretation related to few-shot learning. Second, two categories of few-shot learning methods, i.e., the data-augmentation-based and the prior-knowledge-based, are introduced for the interpretation of remote sensing images. Then, three typical remote sensing interpretation applications are listed, including scene classification, semantic segmentation, and object detection, together with the corresponding public datasets and the evaluation criteria. Finally, the research status is summarized, and some possible research directions are provided. This article gives a reference for scholars working on few-shot learning research in the remote sensing field.

/pdf/research-progress-on-few-shot-learning-for-remote-sensing-579ow3nd5w.pdf

Research Progress on Few-Shot Learning for Remote Sensing Image Interpretation

Convolutional neural networks (CNNs) are becoming increasingly popular in modern remote sensing image processing tasks and exhibit outstanding capability for hyperspectral image (HSI) classification. However, for the existing CNN-based HSI-classification methods, most of them only consider single-scale feature extraction, which may neglect some important fine information and cannot guarantee to capture optimal spatial features. Moreover, many state-of-the-art methods have a huge number of network parameters needed to be tuned, which will cause high computational cost. To address the aforementioned two issues, a novel multiscale residual network (MSRN) is proposed for HSI classification. Specifically, the proposed MSRN introduces depthwise separable convolution (DSC) and replaces the ordinary depthwise convolution in DSC with mixed depthwise convolution (MDConv), which mixes up multiple kernel sizes in a single depthwise convolution operation. The DSC with mixed depthwise convolution (MDSConv) can not only explore features at different scales from each feature map but also greatly reduce learnable parameters in the network. In addition, a multiscale residual block (MRB) is designed by replacing the convolutional layer in an ordinary residual block with the MDSConv layer. The MRB is used as the major unit of the proposed MSRN. Furthermore, to enhance further the feature representation ability, the proposed network adds a high-level shortcut connection (HSC) on the cascaded two MRBs to aggregate lower level features and higher level features. Experimental results on three benchmark HSIs demonstrate the superiority of the proposed MSRN method over several state-of-the-art methods.

Multiscale Residual Network With Mixed Depthwise Convolution for Hyperspectral Image Classification

With the development of high-resolution optical sensors, the classification of ground objects combined with multivariate optical sensors is a hot topic at present. Deep learning methods, such as convolutional neural networks, are applied to feature extraction and classification. In this work, a novel deep belief network (DBN) hyperspectral image classification method based on multivariate optical sensors and stacked by restricted Boltzmann machines is proposed. We introduced the DBN framework to classify spatial hyperspectral sensor data on the basis of DBN. Then, the improved method (combination of spectral and spatial information) was verified. After unsupervised pretraining and supervised fine-tuning, the DBN model could successfully learn features. Additionally, we added a logistic regression layer that could classify the hyperspectral images. Moreover, the proposed training method, which fuses spectral and spatial information, was tested over the Indian Pines and Pavia University datasets. The advantages of this method over traditional methods are as follows: (1) the network has deep structure and the ability of feature extraction is stronger than traditional classifiers; (2) experimental results indicate that our method outperforms traditional classification and other deep learning approaches.

https://www.mdpi.com/1424-8220/19/1/204/pdf

Deep Belief Network for Spectral⁻Spatial Classification of Hyperspectral Remote Sensor Data.

In this paper, we consider building extraction from high spatial resolution remote sensing images. At present, most building extraction methods are based on artificial features. However, the diversity and complexity of buildings mean that building extraction methods still face great challenges, so methods based on deep learning have recently been proposed. In this paper, a building extraction framework based on a convolution neural network and edge detection algorithm is proposed. The method is called Mask R-CNN Fusion Sobel. Because of the outstanding achievement of Mask R-CNN in the field of image segmentation, this paper improves it and then applies it in remote sensing image building extraction. Our method consists of three parts. First, the convolutional neural network is used for rough location and pixel level classification, and the problem of false and missed extraction is solved by automatically discovering semantic features. Second, Sobel edge detection algorithm is used to segment building edges accurately so as to solve the problem of edge extraction and the integrity of the object of deep convolutional neural networks in semantic segmentation. Third, buildings are extracted by the fusion algorithm. We utilize the proposed framework to extract the building in high-resolution remote sensing images from Chinese satellite GF-2, and the experiments show that the average value of IOU (intersection over union) of the proposed method was 88.7% and the average value of Kappa was 87.8%, respectively. Therefore, our method can be applied to the recognition and segmentation of complex buildings and is superior to the classical method in accuracy.

An Efficient Building Extraction Method from High Spatial Resolution Remote Sensing Images Based on Improved Mask R-CNN.

Hyperspectral remote sensing image (HSI) has the characteristics of large data volume and high spectral resolution. It contains abundant spectral information and has tremendous applicable value. Convolutional neural network (CNN) has been successfully applied to HSI classification. However, the limited labeled samples of the HSI make the existing CNN based HSI classification methods generally be plagued by small sample size problem and class imbalance, which cause great challenges for HSI classification. This work proposes a novel CNN architecture for HSI classification. The proposed CNN is a multi-branch fusion network, which is formed by merging multiple branches on an ordinary CNN. It can effectively extract features of HSIs. In addition, the 1 × 1 convolutional layer is introduced into the branches to reduce the number of parameters and then improve the classification efficiency. Furthermore, the L2 regularization is introduced into this work to improve the generalization performance of the proposed model under small sample set. Experimental results on three benchmark hyperspectral images demonstrate that the proposed CNN can provide excellent classification performance under small training set.

Multi-branch fusion network for hyperspectral image classification

Recently, convolutional neural networks (CNNs) have been introduced for hyperspectral image (HSI) classification and shown considerable classification performance. However, the previous CNNs designed for spectral-spatial HSI classification lay stress on the learning for the spatial correlation of HSI data and neglect the channel responses of feature maps. Furthermore, the lack of training samples remains the major challenge for CNN-based HSI classification methods to achieve better performance. To address the aforementioned issues, this paper proposes a new end-to-end pre-activation residual attention network (PRAN) for HSI classification. The pre-activation mechanism and attention mechanism are introduced into the proposed network, and a pre-activation residual attention block (PRAB) is designed, which allows the proposed network to carry adaptively feature recalibration of channel responses and learn more robust spectral-spatial joint feature representations. The proposed PRAN is equipped with two PRABs and several convolutional layers with different kernel sizes, which enables the PRAN to extract high-level discriminative features. Experimental results on three benchmark HSI datasets reveal that the proposed method is provided with competitive performance over several state-of-the-art HSI classification methods, especially when the training set size is relatively small.

/pdf/hyperspectral-image-classification-with-pre-activation-3au8ktm6jm.pdf

Hongmin Gao

Papers

Multiscale Residual Network With Mixed Depthwise Convolution for Hyperspectral Image Classification

Deep Belief Network for Spectral⁻Spatial Classification of Hyperspectral Remote Sensor Data.

An Efficient Building Extraction Method from High Spatial Resolution Remote Sensing Images Based on Improved Mask R-CNN.

Multi-branch fusion network for hyperspectral image classification

Hyperspectral Image Classification With Pre-Activation Residual Attention Network