Crop classification in agriculture is one of important applications for polarimetric synthetic aperture radar (PolSAR) data. For agricultural crop discrimination, compared with single-temporal data, multi-temporal data can dramatically increase crop classification accuracies since the same crop shows different external phenomena as it grows up. In practice, the utilization of multi-temporal data encounters a serious problem known as a “dimension disaster”. Aiming to solve this problem and raise the classification accuracy, this study developed a feature dimension reduction method using stacked sparse auto-encoders (S-SAEs) for crop classification. First, various incoherent scattering decomposition algorithms were employed to extract a variety of detailed and quantitative parameters from multi-temporal PolSAR data. Second, based on analyzing the configuration and main parameters for constructing an S-SAE, a three-hidden-layer S-SAE network was built to reduce the dimensionality and extract effective features to manage the “dimension disaster” caused by excessive scattering parameters, especially for multi-temporal, quad-pol SAR images. Third, a convolutional neural network (CNN) was constructed and employed to further enhance the crop classification performance. Finally, the performances of the proposed strategy were assessed with the simulated multi-temporal Sentinel-1 data for two experimental sites established by the European Space Agency (ESA). The experimental results showed that the overall accuracy with the proposed method was raised by at least 17% compared with the long short-term memory (LSTM) method in the case of a 1% training ratio. Meanwhile, for a CNN classifier, the overall accuracy was almost 4% higher than those of the principle component analysis (PCA) and locally linear embedded (LLE) methods. The comparison studies clearly demonstrated the advantage of the proposed multi-temporal crop classification methodology in terms of classification accuracy, even with small training ratios.

Convolutional neural network (CNN) models have recently demonstrated impressive classification and recognition performance on image and video processing scope. In this paper, we investigate the application of CNN to identifying modulation classes for digitally modulated signals. First, the received baseband data samples of modulated signal are gathered up and transformed to generate the constellation-like training images for convolutional networks. Among the resulting training images, the proposed convolutional gray image is preferred for network training and inference because of the lower computational burden. Second, we propose to use a multiple-scale convolutional neural network (MSCNN) as the classifier. The skip-connection technique is deployed for mitigating the negative effect of vanishing gradients and overfitting during the network training process. Numerical simulations have been carried out to validate the effectiveness of the proposed scheme, the results show that the proposed scheme outperforms the traditional algorithms in terms of classification accuracy.

As an essential task in remote sensing, hyperspectral image classification (HSIC) has been extensively studied. Although many methods based on deep learning have been inducted to improve the performance of HSIC, they still face two challenges: 1) Can the joint spectral–spatial information in the hyperspectral image (HSI) be effectively utilized? 2) Does each pixel in a certain sample contribute equally to classification? We exploit a spectral–spatial attention tensor network (SSATN) in this letter, where the coordinate attention (CA) mechanism is first introduced into a full tensor network. We present a learnable tensor squeezing projection layer (TSPL) instead of the classical pooling layer for the CA block, which enables the network to selectively focus on discriminative features in spectral and spatial. Besides, the SSATN accepts raw HSI data as the input without dimensionality reduction in advance. Compared with the state-of-the-art convolutional neural network (CNN) methods, SSATN can effectively capture certain spectral–spatial features via tensor transformation with fewer parameters. The experimental results on two widely used HSI datasets prove the advantage of the SSATN method.

Nowadays, foundation models have demonstrated exceptional performance across numerous downstream tasks. However, the effective application of these models to hyperspectral image classification (HSIC) is challenged by the unique characteristics of hyperspectral data, including high dimensionality, high variability, and high spatial structure complexity. Therefore, methods need to be developed that leverage the advantages of foundation models while addressing these challenges. Firstly, a novel HSIC algorithm based on a frozen-parameter Segment Anything Model (SAM) encoder, called SAGFFNet, is proposed. This framework represents the first attempt to use a frozen SAM encoder for global feature extraction and to use spectral dimension data as prompts, enabling precise global spatial-spectral feature extraction with the aid of spectral information. Secondly, by introducing the self-adaptive padding mechanism and the global feature extraction subnetwork (GFEsNet), the model is enabled to extract distinctive and discriminative features for each category from hyperspectral data through varying padding sizes, thereby enhancing the feature extraction and generalization capabilities of the foundation model. Subsequently, the spectral feature prompt subnetwork (SFPsNet) is designed to extract spectral feature information from samples of different classes as prompt features, assisting the framework in better understanding the global features extracted by GFEsNet. Finally, the semantic information decoder subnetwork (SIDsNet) is introduced as a semantic information decoder, achieving efficient fusion of global spatial-spectral features and spectral prompt features, which significantly improves classification performance. Experiments conducted on four hyperspectral image datasets show that the proposed method outperforms nine existing approaches in terms of classification accuracy.

Accurate and reliable crop classification information is a significant data source for agricultural monitoring and food security evaluation research. It is well-known that polarimetric synthetic aperture radar (PolSAR) data provides ample information for crop classification. Moreover, multi-temporal PolSAR data can further increase classification accuracies since the crops show different external forms as they grow up. In this paper, we distinguish the crop types with multi-temporal PolSAR data. First, due to the “dimension disaster” of multi-temporal PolSAR data caused by excessive scattering parameters, a neural network of sparse auto-encoder with non-negativity constraint (NC-SAE) was employed to compress the data, yielding efficient features for accurate classification. Second, a novel crop discrimination network with multi-scale features (MSCDN) was constructed to improve the classification performance, which is proved to be superior to the popular classifiers of convolutional neural networks (CNN) and support vector machine (SVM). The performances of the proposed method were evaluated and compared with the traditional methods by using simulated Sentinel-1 data provided by European Space Agency (ESA). For the final classification results of the proposed method, its overall accuracy and kappa coefficient reaches 99.33% and 99.19%, respectively, which were almost 5% and 6% higher than the CNN method. The classification results indicate that the proposed methodology is promising for practical use in agricultural applications.