Attention-Based Spatial and Spectral Network with PCA-Guided Self-Supervised Feature Extraction for Change Detection in Hyperspectral Images

Wang, Zhao; Jiang, Fenlong; Liu, Tongfei; Xie, Fei

doi:10.3390/rs13234927

Cited by 13 publications

(6 citation statements)

References 60 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In recent years, attention mechanisms have been widely used in deep learning [9,[49][50][51], especially computer vision. Attention mechanisms commonly used in computer vision and remote sensing image processing can be divided into two major categories according to the function of the attention mechanism [52,53]: channel attention and spatial attention.…”

Section: Related Workmentioning

confidence: 99%

“…With the development of satellite, aviation, and unmanned aerial vehicle (UAV) technology, huge amounts of high-resolution (HR) remote sensing images have been captured in a constant stream [1][2][3]. These HR remote sensing images have been applied to land cover classification [4][5][6], change detection [7][8][9], target recognition [10,11], and image restoration and registration [12,13], for example. This brings opportunities for us to observe fine objects such as buildings, roads, vehicles, etc.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Frequency Spectrum Intensity Attention Network for Building Detection from High-Resolution Imagery

2022

View full text Add to dashboard Cite

Computational intelligence techniques have been widely used for automatic building detection from high-resolution remote sensing imagery and especially the methods based on neural networks. However, existing methods do not pay attention to the value of high-frequency and low-frequency information in the frequency domain for feature extraction of buildings in remote sensing images. To overcome these limitations, this paper proposes a frequency spectrum intensity attention network (FSIANet) with an encoder–decoder structure for automatic building detection. The proposed FSIANet mainly involves two innovations. One, a novel and plug-and-play frequency spectrum intensity attention (FSIA) mechanism is devised to enhance feature representation by evaluating the informative abundance of the feature maps. The FSIA is deployed after each convolutional block in the proposed FSIANet. Two, an atrous frequency spectrum attention pyramid (AFSAP) is constructed by introducing FSIA in widely used atrous spatial pyramid pooling. The AFSAP is able to select the features with high response to building semantic features at each scale and weaken the features with low response, thus enhancing the feature representation of buildings. The proposed FSIANet is evaluated on two large public datasets (East Asia and Inria Aerial Image Dataset), which demonstrates that the proposed method can achieve the state-of-the-art performance in terms of F1-score and intersection-over-union.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Frequency Spectrum Intensity Attention Network for Building Detection from High-Resolution Imagery

2022

View full text Add to dashboard Cite

show abstract

“…In the application field of hyperspectral images, classification is a very important processing technology, which can provide a basis for recognition, target detection, etc. Hyperspectral image classification mainly uses spectral features for classification, and spectral feature extraction is divided into unsupervised feature extraction method [7] and supervised feature extraction method [8], according to whether it depends on the existence of labeled samples.…”

Section: Introductionmentioning

confidence: 99%

Hyperspectral image classification based on spectral feature extraction

Wang

2023

International Conference on Image, Signal Processing, and Pattern Recognition (ISPP 2023)

View full text Add to dashboard Cite

Hyperspectral remote sensing can synchronously obtain the surface coverage space image and continuous spectral data, and can realize fine classification and recognition of ground objects. The motivation for this paper is to search an effective specteal feature extraction method for hyperspectral classification. The main methods used to extract spectral features of hyperspectral images are principal component analysis (PCA) and linear discriminant analysis (LDA). The realization of classification based on k-nearest neighbor (KNN) classifier is discussed. The training set and test set of Pavia University, Indian Pines and Salinas are selected as the data source, and the selection experiment of algorithm parameters is set, and the classification performance under different feature extraction methods is compared and analyzed. The experimental results show that with the parameters of OA, PA and Kappa based on the confusion matrix as the evaluation indicators, the classification accuracy based on LDA method is higher, and the effective classification of ground objects can be achieved. The results could provide a basis for further research on hyperspectral image classification based on spatial spectral feature extraction.

show abstract

“…Gong et al [42] incorporated spectral and spatial attention mechanisms to selectively weight the various bands and regions in the input images for CD. Wang et al [43] introduced a simple attention mechanism to measure the weights of different features before concatenating them. Huang et al [44] integrated parallel spatial and spectral attention to adaptively enhance the relevant global dependencies.…”

Section: Introductionmentioning

confidence: 99%

Spectral–Temporal Transformer for Hyperspectral Image Change Detection

Ding

2023

Remote Sensing

View full text Add to dashboard Cite

Deep-Learning-based (DL-based) approaches have achieved remarkable performance in hyperspectral image (HSI) change detection (CD). Convolutional Neural Networks (CNNs) are often employed to capture fine spatial features, but they do not effectively exploit the spectral sequence information. Furthermore, existing Siamese-based networks ignore the interaction of change information during feature extraction. To address this issue, we propose a novel architecture, the Spectral–Temporal Transformer (STT), which processes the HSI CD task from a completely sequential perspective. The STT concatenates feature embeddings in spectral order, establishing a global spectrum–time-receptive field that can learn different representative features between two bands regardless of spectral or temporal distance, thereby strengthening the learning of temporal change information. Via the multi-head self-attention mechanism, the STT is capable of capturing spectral–temporal features that are weighted and enriched with discriminative sequence information, such as inter-spectral correlations, variations, and time dependency. We conducted experiments on three HSI datasets, demonstrating the competitive performance of our proposed method. Specifically, the overall accuracy of the STT outperforms the second-best method by 0.08%, 0.68%, and 0.99% on the Farmland, Hermiston, and River datasets, respectively.

show abstract

Attention-Based Spatial and Spectral Network with PCA-Guided Self-Supervised Feature Extraction for Change Detection in Hyperspectral Images

Cited by 13 publications

References 60 publications

Frequency Spectrum Intensity Attention Network for Building Detection from High-Resolution Imagery

Frequency Spectrum Intensity Attention Network for Building Detection from High-Resolution Imagery

Hyperspectral image classification based on spectral feature extraction

Spectral–Temporal Transformer for Hyperspectral Image Change Detection

Contact Info

Product

Resources

About