This paper proposes a novel deep learning framework named bidirectional-convolutional long short term memory (Bi-CLSTM) network to automatically learn the spectral-spatial features from hyperspectral images (HSIs). In the network, the issue of spectral feature extraction is considered as a sequence learning problem, and a recurrent connection operator across the spectral domain is used to address it. Meanwhile, inspired from the widely used convolutional neural network (CNN), a convolution operator across the spatial domain is incorporated into the network to extract the spatial feature. In addition, to sufficiently capture the spectral information, a bidirectional recurrent connection is proposed. In the classification phase, the learned features are concatenated into a vector and fed to a Softmax classifier via a fully-connected operator. To validate the effectiveness of the proposed Bi-CLSTM framework, we compare it with six state-of-the-art methods, including the popular 3D-CNN model, on three widely used HSIs (i.e., Indian Pines, Pavia University, and Kennedy Space Center). The obtained results show that Bi-CLSTM can improve the classification performance by almost as compared to 3D-CNN.
Deep learning has shown its huge potential in the field of hyperspectral image (HSI) classification. However, most of the deep learning models heavily depend on the quantity of available training samples. In this article, we propose a multitask generative adversarial network (MTGAN) to alleviate this issue by taking advantage of the rich information from unlabeled samples. Specifically, we design a generator network to simultaneously undertake two tasks: the reconstruction task and the classification task. The former task aims at reconstructing an input hyperspectral cube, including the labeled and unlabeled ones, whereas the latter task attempts to recognize the category of the cube. Meanwhile, we construct a discriminator network to discriminate the input sample coming from the real distribution or the reconstructed one. Through an adversarial learning method, the generator network will produce real-like cubes, thus indirectly improving the discrimination and generalization ability of the classification task. More importantly, in order to fully explore the useful information from shallow layers, we adopt skip-layer connections in both reconstruction and classification tasks. The proposed MTGAN model is implemented on three standard HSIs, and the experimental results show that it is able to achieve higher performance than other state-of-the-art deep learning models.
8 May 2019 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Low spatial resolution hyperspectral (LRHS) and high spatial resolution multispectral (HRMS) image fusion has been recognized as an important technology for enhancing the spatial resolution of LRHS image. Recent advances in convolutional neural network have improved the performance of state-of-the-art fusion methods. However, it is still a challenging problem to effectively explore the spatial information of HRMS image. In this paper, we propose a pyramid fully convolutional network made up of an encoder sub-network and a pyramid fusion sub-network to address this issue. Specifically, the encoder sub-network aims to encode the LRHS image into a latent image. Then, this latent image, together with a HRMS image pyramid input, is used to progressively reconstruct the high spatial resolution hyperspectral image in a global-to-local manner. Furthermore, to sharpen the blurry predictions easily obtained by the standard $l_{2}$ loss, we introduce the gradient difference loss as a regularization term. We evaluate the proposed method on three datasets acquired by three different satellite sensors. Experimental results demonstrate that the proposed method achieves better performance than several state-of-the-art methods.
16 September 2022 in IEEE Transactions on Geoscience and Remote Sensing
Semantic segmentation of high-resolution remote sensing imageries (HRSIs) is a critical task for a wide range of applications, such as precision agriculture and urban planning. Although convolutional neural networks (CNNs) have made great progress in accomplishing this task recently, there still exist some challenges to address, one of which is simultaneously segmenting objects with large-scale variations in a HRSI. Targeting at this challenge, previous CNNs often adopt multiple convolution kernels in one layer or skip-layer connections between different layers to extract multiscale representations. However, due to the limited learning capacity of each CNN, it tends to make tradeoffs in segmenting different-scale objects. This would lead to unsatisfactory segmentation results for some objects, especially the small or the large ones. In this article, we propose a multiscale progressive segmentation network to address this issue. Instead of forcing one network to deal with all scales of objects, our network attempts to cascade three subnetworks for gradually segmenting objects into small scale, large scale, and other scale. In order to make the subnetwork focus on the specific scale objects, a scale guidance module is designed. It takes advantage of segmentation results from the preceding subnetwork to guide the feature learning of the succeeding one. Additionally, to acquire the final segmentation results, we propose a position-sensitive module for adaptively combining the outputs of the three subnetworks. This module is capable of assigning combination weights of different subnetworks according to their importance. Experiments on two benchmark datasets named Vaihingen and Potsdam indicate that our proposed network can achieve considerable improvements in comparison with several state-of-the-art segmentation models.
15 July 2020 in IEEE Transactions on Geoscience and Remote Sensing
Contextual information has been demonstrated to be helpful for airborne image segmentation. However, most of the previous works focus on the exploitation of spatially contextual information, which is difficult to segment isolated objects, mainly surrounded by uncorrelated objects. To alleviate this issue, we attempt to take advantage of the co-occurrence relations between different classes of objects in the scene. Especially, similar to other works, convolutional features are first extracted to capture the spatially contextual information. Then, a feature decoupling module is designed to encode the class co-occurrence relations into the convolutional features; thus, the most discriminative features can be decoupled. Finally, the segmentation result is inferred from the decoupled features. The whole process is integrated to form an end-to-end network, named class-guided feature decoupling network (CGFDN). Experimental results on two widely used benchmark data sets show that CGFDN obtains competitive results (>90% overall accuracy (OA) on 5-cm-resolution Potsdam and >91% OA on 9-cm-resolution Vaihingen) in comparison with several state-of-the-art models.