A Comprehensive Survey of Deep Learning in Remote Sensing: Theories, Tools and Challenges for the Community
TL;DR: This work focuses on theories, tools, and challenges for the RS community, and focuses on unsolved challenges and opportunities as they relate to inadequate data sets, big data, and human-understandable solutions for modeling physical phenomena.
Abstract: In recent years, deep learning (DL), a re-branding of neural networks (NNs), has risen to the top in numerous areas, namely computer vision (CV), speech recognition, natural language processing, etc. Whereas remote sensing (RS) possesses a number of unique challenges, primarily related to sensors and applications, inevitably RS draws from many of the same theories as CV; e.g., statistics, fusion, and machine learning, to name a few. This means that the RS community should be aware of, if not at the leading edge of, of advancements like DL. Herein, we provide the most comprehensive survey of state-of-the-art RS DL research. We also review recent new developments in the DL field that can be used in DL for RS. Namely, we focus on theories, tools and challenges for the RS community. Specifically, we focus on unsolved challenges and opportunities as it relates to (i) inadequate data sets, (ii) human-understandable solutions for modelling physical phenomena, (iii) Big Data, (iv) non-traditional heterogeneous data sources, (v) DL architectures and learning algorithms for spectral, spatial and temporal data, (vi) transfer learning, (vii) an improved theoretical understanding of DL systems, (viii) high barriers to entry, and (ix) training and optimizing the DL.
Citations
More filters
TL;DR: 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 is provided, providing an exhaustive comparison of the discussed techniques.
Abstract: 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 .
534 citations
Cites background or methods from "A Comprehensive Survey of Deep Lear..."
...Contextualizing the evolution of DL methods, at early days, neural models emerged within the fields of pattern recognition and signal processing, inspired by the behaviour of the biological brain and implementing a hierarchical structure where each part of the stack conforms a layer, being neurons (also perceptrons) the basic unit of each layer (Ball et al., 2017)....
[...]
...DBNs combine probability and graph theory to implement a generative probabilistic graphical model (PGM) with the structure of a directed acyclic graph (DAG) (Ball et al., 2017)....
[...]
...These coupled networks are trained together as an end-to-end model to optimize all the weights in the CNN: (i) the FE-net, composed by a hierarchical stack of feature extraction and detection stages that learns high-level representations of the inputs, and (ii) the classifier, composed by a stack of FC layers that performs the final classification task, computing the membership of each input sample to a certain class (Ball et al., 2017)....
[...]
...erative probabilistic graphical model (PGM) with the structure of a directed acyclic graph (DAG) (Ball et al., 2017)....
[...]
TL;DR: It is indicated that multimodal data fusion using low-cost UAV within a DNN framework can provide a relatively accurate and robust estimation of crop yield, and deliver valuable insight for high-throughput phenotyping and crop field management with high spatial precision.
362 citations
TL;DR: A detailed investigation of state-of-the-art deep learning tools for classification of complex wetland classes using multispectral RapidEye optical imagery for wetland mapping in Canada finds InceptionResNetV2 is consistently found to be superior compared to all other convnets, suggesting the integration of Inception and ResNet modules is an efficient architecture for classifying complex remote sensing scenes such as wetlands.
Abstract: Despite recent advances of deep Convolutional Neural Networks (CNNs) in various computer vision tasks, their potential for classification of multispectral remote sensing images has not been thoroughly explored. In particular, the applications of deep CNNs using optical remote sensing data have focused on the classification of very high-resolution aerial and satellite data, owing to the similarity of these data to the large datasets in computer vision. Accordingly, this study presents a detailed investigation of state-of-the-art deep learning tools for classification of complex wetland classes using multispectral RapidEye optical imagery. Specifically, we examine the capacity of seven well-known deep convnets, namely DenseNet121, InceptionV3, VGG16, VGG19, Xception, ResNet50, and InceptionResNetV2, for wetland mapping in Canada. In addition, the classification results obtained from deep CNNs are compared with those based on conventional machine learning tools, including Random Forest and Support Vector Machine, to further evaluate the efficiency of the former to classify wetlands. The results illustrate that the full-training of convnets using five spectral bands outperforms the other strategies for all convnets. InceptionResNetV2, ResNet50, and Xception are distinguished as the top three convnets, providing state-of-the-art classification accuracies of 96.17%, 94.81%, and 93.57%, respectively. The classification accuracies obtained using Support Vector Machine (SVM) and Random Forest (RF) are 74.89% and 76.08%, respectively, considerably inferior relative to CNNs. Importantly, InceptionResNetV2 is consistently found to be superior compared to all other convnets, suggesting the integration of Inception and ResNet modules is an efficient architecture for classifying complex remote sensing scenes such as wetlands.
296 citations
Journal Article•
TL;DR: This result is proved here for a class of nodes termed "semi-algebraic gates" which includes the common choices of ReLU, maximum, indicator, and piecewise polynomial functions, therefore establishing benefits of depth against not just standard networks with ReLU gates, but also convolutional networks with reLU and maximization gates, sum-product networks, and boosted decision trees.
Abstract: For any positive integer $k$, there exist neural networks with $\Theta(k^3)$ layers, $\Theta(1)$ nodes per layer, and $\Theta(1)$ distinct parameters which can not be approximated by networks with $\mathcal{O}(k)$ layers unless they are exponentially large --- they must possess $\Omega(2^k)$ nodes. This result is proved here for a class of nodes termed "semi-algebraic gates" which includes the common choices of ReLU, maximum, indicator, and piecewise polynomial functions, therefore establishing benefits of depth against not just standard networks with ReLU gates, but also convolutional networks with ReLU and maximization gates, sum-product networks, and boosted decision trees (in this last case with a stronger separation: $\Omega(2^{k^3})$ total tree nodes are required).
288 citations
References
More filters
TL;DR: A novel, efficient, gradient based method called long short-term memory (LSTM) is introduced, which can learn to bridge minimal time lags in excess of 1000 discrete-time steps by enforcing constant error flow through constant error carousels within special units.
Abstract: Learning to store information over extended time intervals by recurrent backpropagation takes a very long time, mostly because of insufficient, decaying error backflow. We briefly review Hochreiter's (1991) analysis of this problem, then address it by introducing a novel, efficient, gradient based method called long short-term memory (LSTM). Truncating the gradient where this does not do harm, LSTM can learn to bridge minimal time lags in excess of 1000 discrete-time steps by enforcing constant error flow through constant error carousels within special units. Multiplicative gate units learn to open and close access to the constant error flow. LSTM is local in space and time; its computational complexity per time step and weight is O. 1. Our experiments with artificial data involve local, distributed, real-valued, and noisy pattern representations. In comparisons with real-time recurrent learning, back propagation through time, recurrent cascade correlation, Elman nets, and neural sequence chunking, LSTM leads to many more successful runs, and learns much faster. LSTM also solves complex, artificial long-time-lag tasks that have never been solved by previous recurrent network algorithms.
72,897 citations
01 Jan 1998
TL;DR: In this article, a graph transformer network (GTN) is proposed for handwritten character recognition, which can be used to synthesize a complex decision surface that can classify high-dimensional patterns, such as handwritten characters.
Abstract: Multilayer neural networks trained with the back-propagation algorithm constitute the best example of a successful gradient based learning technique. Given an appropriate network architecture, gradient-based learning algorithms can be used to synthesize a complex decision surface that can classify high-dimensional patterns, such as handwritten characters, with minimal preprocessing. This paper reviews various methods applied to handwritten character recognition and compares them on a standard handwritten digit recognition task. Convolutional neural networks, which are specifically designed to deal with the variability of 2D shapes, are shown to outperform all other techniques. Real-life document recognition systems are composed of multiple modules including field extraction, segmentation recognition, and language modeling. A new learning paradigm, called graph transformer networks (GTN), allows such multimodule systems to be trained globally using gradient-based methods so as to minimize an overall performance measure. Two systems for online handwriting recognition are described. Experiments demonstrate the advantage of global training, and the flexibility of graph transformer networks. A graph transformer network for reading a bank cheque is also described. It uses convolutional neural network character recognizers combined with global training techniques to provide record accuracy on business and personal cheques. It is deployed commercially and reads several million cheques per day.
42,067 citations
07 Jun 2015
TL;DR: Inception as mentioned in this paper is a deep convolutional neural network architecture that achieves the new state of the art for classification and detection in the ImageNet Large-Scale Visual Recognition Challenge 2014 (ILSVRC14).
Abstract: We propose a deep convolutional neural network architecture codenamed Inception that achieves the new state of the art for classification and detection in the ImageNet Large-Scale Visual Recognition Challenge 2014 (ILSVRC14). The main hallmark of this architecture is the improved utilization of the computing resources inside the network. By a carefully crafted design, we increased the depth and width of the network while keeping the computational budget constant. To optimize quality, the architectural decisions were based on the Hebbian principle and the intuition of multi-scale processing. One particular incarnation used in our submission for ILSVRC14 is called GoogLeNet, a 22 layers deep network, the quality of which is assessed in the context of classification and detection.
40,257 citations
TL;DR: The ImageNet Large Scale Visual Recognition Challenge (ILSVRC) as mentioned in this paper is a benchmark in object category classification and detection on hundreds of object categories and millions of images, which has been run annually from 2010 to present, attracting participation from more than fifty institutions.
Abstract: The ImageNet Large Scale Visual Recognition Challenge is a benchmark in object category classification and detection on hundreds of object categories and millions of images. The challenge has been run annually from 2010 to present, attracting participation from more than fifty institutions. This paper describes the creation of this benchmark dataset and the advances in object recognition that have been possible as a result. We discuss the challenges of collecting large-scale ground truth annotation, highlight key breakthroughs in categorical object recognition, provide a detailed analysis of the current state of the field of large-scale image classification and object detection, and compare the state-of-the-art computer vision accuracy with human accuracy. We conclude with lessons learned in the 5 years of the challenge, and propose future directions and improvements.
30,811 citations
TL;DR: The relationship between transfer learning and other related machine learning techniques such as domain adaptation, multitask learning and sample selection bias, as well as covariate shift are discussed.
Abstract: A major assumption in many machine learning and data mining algorithms is that the training and future data must be in the same feature space and have the same distribution. However, in many real-world applications, this assumption may not hold. For example, we sometimes have a classification task in one domain of interest, but we only have sufficient training data in another domain of interest, where the latter data may be in a different feature space or follow a different data distribution. In such cases, knowledge transfer, if done successfully, would greatly improve the performance of learning by avoiding much expensive data-labeling efforts. In recent years, transfer learning has emerged as a new learning framework to address this problem. This survey focuses on categorizing and reviewing the current progress on transfer learning for classification, regression, and clustering problems. In this survey, we discuss the relationship between transfer learning and other related machine learning techniques such as domain adaptation, multitask learning and sample selection bias, as well as covariate shift. We also explore some potential future issues in transfer learning research.
18,616 citations