In recent years, active learning (AL) methods have provided a feasible approach to alleviate the problem of limited labeled samples in deep learning projects. Existing AL algorithms generally tend to select sample without labeled, whose category is difficult to distinguish. However, the sample in the category center is difficult to determine in AL operations, resulting in inaccurate category measuring and inaccurate sample selection. In addition, hyperspectral images (HSIs) have rich spectral reflective bands with strong correlations, which leads to the phenomenon that the spatial distribution between different categories in HSIs characterizes staggered distribution, which undoubtedly influences the HSI classification effect. In this article, we propose a new AL method (called PLGCN) which combines prototype learning (PL) and graph convolution network (GCN) to solve few-shot HSI classification tasks, and this method can add into existing deep learning-based HSI classification models. It includes two advantages: 1) the prototype of each category is iteratively updated to ensure the optimality of prototype in each sampling stage and 2) the spatial distribution of unlabeled samples is extracted via graph convolution neural network in order to obtain the better features in new space for easier discriminating. Experimental results on three commonly used benchmark HSI datasets demonstrate the effectiveness of the PLGCN in HSI classification tasks with limited labeled samples.

In the world, with the continuous development of modern society and the acceleration of urbanization, the problem of air pollution is becoming increasingly salient. Methods for predicting the air quality grade and determining the necessary governance are at present most urgent problems waiting to be solved by human beings. In recent years, more and more machine-learning-based methods have been used to solve the air quality prediction problem. However, the uncertainty of environmental changes and the difficulty of precisely predicting quantitative values seriously influence prediction results. In this paper, the proposed air pollutant quality grade prediction method based on a hyperparameter-optimization-inspired long short-term memory (LSTM) network provides two advantages. Firstly, the definition of air quality grade is introduced in the air quality prediction task, which turns a fitting problem into a classification problem and makes the complex problem simple; secondly, the hunter–prey optimization algorithm is used to optimize the hyperparameters of the LSTM structure to obtain the optimal network structure adaptively determined through the use of input data, which can include more generalization abilities. The experimental results from three real Xi’an air quality datasets display the effectiveness of the proposed method.

Convolutional neural networks (CNNs) have been widely used in the field of hyperspectral image change detection (HSI-CD) due to their excellent feature extraction capabilities. However, traditional convolution methods usually rely on a limited number of convolution kernels when extracting local features, and the local information only covers a small range of details, making it difficult to flexibly extract local changes at different levels in the image. In addition, CNNs have certain limitations in modeling global dependencies. In recent years, state-space model (SSM)-based methods (such as mamba) have effectively solved the quadratic complexity problem of transformer in capturing long-distance global dependencies. However, due to the processing method of expanding images into sequences, the spatial location structure of the image is destroyed, the spatial correlation between adjacent pixels is weakened, and spatial location information is easily lost, resulting in poor performance on complex hyperspectral datasets. To address these issues, we propose a wavelet-augmented dual-branch position-embedding mamba network (WDP-Mamba) for hyperspectral image change detection. This network combines the complementary advantages of mamba and wavelet convolution (WTconv) in global feature extraction and local edge feature extraction. Specifically, our method includes the following key innovations: 1) a novel state-space model, the adaptive position residual state space block (APRSSB) is designed, embedding learnable positional encodings within the mamba framework. Through a dual-temporal spatial-spectral scoring mechanism, the model dynamically preserves the topological relationships of pixel positions, enabling global dependency modeling with linear complexity; 2) a multi-level frequency-domain aware (MLFA) module is proposed to fully extract representative local features by decomposing information into frequency components of different scales and analyzing signals locally in multiple time and frequency windows. Extensive experiments on three hyperspectral datasets show that WDP-Mamba outperforms existing methods in both qualitative and quantitative assessments. Compared with other comparative models, WDP-Mamba achieves significant improvements in multiple performance metrics with a lower number of parameters, demonstrating its superior performance in the HSI-CD task.

Cross-domain few-shot learning (FSL) has demonstrated remarkable new classes recognition capabilities in hyperspectral image classification tasks. However, existing domain adaptation methods face two critical challenges in the cross-domain feature alignment process: first, the domain shift leads to misaligned feature transfer and diminished classification accuracy; second, the intra-class feature dispersion and inter-class boundary blurring in few-shot tasks result in degraded classification performance for novel classes. Moreover, the impact of redundant and noisy data on model discriminability is rarely considered in existing approaches. To solve these issues, this article proposes a cross-domain FSL hyperspectral image classification method based on diffusion-augmented prototype knowledge distillation (DAPKD-CFSL). Firstly, we introduce a diffusion-augmented unsupervised domain adaptation pre-training (DA-PT) framework to address the domain shift by performing a domain-adversarial denoising and reconstruction task using visible source data and masked target data. Second, our dual-branch spatial-spectral attention (DB-SSA) captures global and local spectral-spatial dependencies to enhance feature representation. Then, the proposed global-local prototype knowledge distillation (GL-PKD) performs global prototype alignment while conducting local contrastive learning, addressing feature dispersion and boundary ambiguity. Finally, a dynamic learning strategy prioritizes feature alignment early and gradually strengthens classification supervision through adaptive loss weights, and incorporates an SNR-enhanced loss to effectively mitigate noise interference. The experimental results on three HSI datasets demonstrate the superiority and effectiveness of the proposed DAPKD-CFSL.

Air pollution is becoming a serious concern with the development of society and urban expansion, and predicting air quality is the most pressing problem for human beings. Recently, more and more machine-learning-based methods are being used to solve the air-quality-prediction problem, and gated recurrent units (GRUs) are a representative method because of their advantage for processing time-series data. However, in the same air-quality-prediction task, different researchers have always designed different structures of the GRU due to their different experiences. Data-adaptively designing a GRU structure has thus become a problem. In this paper, we propose an adaptive GRU to address this problem, and the adaptive GRU structures are determined by the dataset, which mainly contributes with three steps. Firstly, an encoding method for the GRU structure is proposed for representing the network structure in a fixed-length binary string; secondly, we define the reciprocal of the sum of the loss of each individual as the fitness function for the iteration computation; thirdly, the genetic algorithm is used for computing the data-adaptive GRU network structure, which can enhance the air-quality-prediction result. The experiment results from three real datasets in Xi’an show that the proposed method achieves better effectiveness in RMSE and SAMPE than the existing LSTM-, SVM-, and RNN-based methods.

In recent years, cross-domain few-shot learning (CDFSL) has demonstrated remarkable performance in hyperspectral image classification (HSIC), partially alleviating the distribution shift problem. However, most domain adaptation methods rely on similarity metrics to establish cross-domain class matching, making it difficult to simultaneously account for intra-class sample size variations and inherent inter-class differences. To address this problem, existing studies have introduced a class weighting mechanism within the prototype network framework, determining class weights by calculating inter-sample similarity through distance metrics. However, this method suffers from a dual limitation: susceptibility to noise interference and insufficient capacity to capture global class variations, which may lead to distorted weight allocation and consequently result in alignment bias. To solve these issues, we propose a novel class-discrepancy dynamic weighting-based cross-domain FSL (CDDW-CFSL) framework. It integrates three key components: (1) the class-weighted domain adaptation (CWDA) method dynamically measures cross-domain distribution shifts using global class mean discrepancies. It employs discrepancy-sensitive weighting to strengthen the alignment of critical categories, enabling accurate domain adaptation while maintaining feature topology; (2) the class mean refinement (CMR) method incorporates class covariance distance to compute distribution discrepancies between support set samples and class prototypes, enabling the precise capture of cross-domain feature internal structures; (3) a novel multi-dimensional feature extractor that captures both local spatial details and continuous spectral characteristics simultaneously, facilitating deep cross-dimensional feature fusion. The results in three publicly available HSIC datasets show the effectiveness of the CDDW-CFSL.