Co-Training has achieved significant success in the field of semi-supervised learning(SSL); however, the homogenization phenomenon, which arises from multiple models tending towards similar decision boundaries, remains inadequately addressed. To tackle this issue, we propose a novel algorithm called β-FFT from the perspectives of data processing and training structure. In data processing, we apply diverse augmentations to input data and feed them into two sub-networks. To balance the training instability caused by different augmentations during consistency learning, we introduce a nonlinear interpolation technique based on the Fast Fourier Transform (FFT). By swapping low-frequency components between variously augmented images, this method not only generates smooth and diverse training samples that bridge different augmentations but also enhances the model’s generalization capability while maintaining consistency learning stability. In training structure, we devise a differentiated training strategy to mitigate homogenization in co-training. Specifically, we use labeled data for additional training of one model within the co-training framework, while for unlabeled data, we employ linear interpolation based on the Beta(β) distribution as a regularization technique in additional training. This approach allows for more efficient utilization of limited labeled data and simultaneously improves the model’s performance on unlabeled data, optimizing overall system performance. Code is available at the following link. https://github.com/Xi-Mu-Yu/beta-FFT.

The localization of Unmanned Aerial Vehicles (UAVs) based on the Global Positioning Navigation System (GNSS) faces potential risks of spoofing and interference. To complete tasks in environments with localization interference or GPS-denied conditions, UAVs need to adopt alternative techniques for self-localization. This paper proposes an absolute visual localization method based on reference maps, named UAV Visual Localization Network (UVL). By performing absolute localization using the learned reference map information, UVL avoids the error accumulation present in relative visual localization methods. The UVL method is designed based on a scene coordinate regression (SCR) paradigm to enable an end-to-end localization solution. Furthermore, UVL integrates a self-attention mechanism based on a visual transformer, which enhances the effective association of scene features. The proposed algorithm was evaluated using the publicly available CrossLoc benchmark datasets and the newly introduced Chang’an Park dataset. The quantitative evaluation results demonstrate that the proposed method achieves superior performance compared to existing approaches.

Recently, few-shot learning-based methods have achieved impressive results in cross-domain hyperspectral classification. However, existing approaches often ignore differences in spectral information caused by varing spectral range across different datasets, and encounter limitations due to the constrained parameter size of the models. Furthermore, the substantial differences between RGB and hyperspectral images present significant challenges in applying foundation models (e.g., SAM, DINOv2) to the hyperspectral domain. This paper proposes a novel framework for cross-domain few-shot hyperspectral classification that leverages parameter-efficient fine-tuning, which we apply to DINOv2 to construct SpectralDINO. Different spectral ranges reflect different physical properties of the target. To enhance the consistency of the spectral features extracted by the model from different domains, we introduce a source domain spectral alignment (SDSA) strategy to align the spectral ranges and bands of the source domain data to the target domain. SpectralDINO employs the visual foundation model to enhance its ability to generalize cross-domain knowledge. Additionally, we propose a dual mixture-of-subspaces low-rank adaptation (Dual-MoS LoRA) method to address the structural limitation of the low-rank adaptation methods in distinguishing domain-specific features from multi-domain inputs. Only 1.14% of the 21.37M parameters need to be trained to perform fine-tuning. Extensive experimental results on three public datasets demonstrate the superiority of SpectralDINO.

Abstract The image‐to‐image translation (I2IT) task aims to transform images from the source domain into the specified target domain. State‐of‐the‐art CycleGAN‐based translation algorithms typically use cycle consistency loss and latent regression loss to constrain translation. In this work, it is demonstrated that the model parameters constrained by the cycle consistency loss and the latent regression loss are equivalent to optimizing the medians of the data distribution and the generative distribution. In addition, there is a style bias in the translation. This bias interacts between the generator and the style encoder and visually exhibits translation errors, e.g. the style of the generated image is not equal to the style of the reference image. To address these issues, a new I2IT model termed high‐quality‐I2IT (HQ‐I2IT) is proposed. The optimization scheme is redesigned to prevent the model from optimizing the median of the data distribution. In addition, by separating the optimization of the generator and the latent code estimator, the redesigned model avoids error interactions and gradually corrects errors during training, thereby avoiding learning the median of the generated distribution. The experimental results demonstrate that the visual quality of the images produced by HQ‐I2IT is significantly improved without changing the generator structure, especially when guided by the reference images. Specifically, the Fréchet inception distance on the AFHQ and CelebA‐HQ datasets are reduced from 19.8 to 10.2 and from 23.8 to 17.0, respectively.

The polarization characteristics of water-leaving radiation, which contain rich information on oceanic constituents, have often been neglected. Due to the lack of suitable instruments and practical difficulties in removing strong contamination by polarized skylight, direct measurement of the polarization of water-leaving radiation remains a challenge. In this study, we designed an above-water instrument (named POLWR) to directly measure the polarization of water-leaving radiation and examined its field application in Qiandao Lake, China. Results showed that the Stokes components of water-leaving radiance ($L_{w}$) measured by POLWR were consistent with the radiative transfer (RT) simulations, with a determination coefficient ($R^{2}$) and mean relative error of 0.67 and 18.86%, respectively. The Qiandao Lake results revealed that the degree of polarization (DOP) of$L_{w}$varied from 0.05 to 0.5 within the 412–865-nm range. Moreover, a good relationship between the polarized remote sensing reflectance ($R_{\mathrm {rsp}}$), and DOP and chlorophyll-a (Chla) concentration was found at 368 nm in this productive lake, indicating great potential for the inversion of oceanic constituents from polarization signals. With its small size and direct measurement ability, the POLWR instrument should be widely applicable and could help improve our understanding of the polarization characteristics of water-leaving radiation and the underwater light field.

Hyperspectral images are mixtures of spectra of materials in a scene. Accurate analysis of hyperspectral image requires spectral unmixing. The result of spectral unmixing is the material spectral signatures and their corresponding fractions. The materials are called endmembers. Endmember extraction equals to acquire spectral signatures of the materials. In this study, the authors propose a new hyperspectral endmember extraction algorithm for hyperspectral image based on QR factorisation using Givens rotations (EEGR). Evaluation of the algorithm is demonstrated by comparing its performance with two popular endmember extraction methods, which are vertex component analysis (VCA) and maximum volume by householder transformation (MVHT). Both simulated mixtures and real hyperspectral image are applied to the three algorithms, and the quantitative analysis of them is presented. EEGR exhibits better performance than VCA and MVHT. Moreover, EEGR algorithm is convenient to implement parallel computing for real‐time applications based on the hardware features of Givens rotations.

Target detection of hyperspectral image (HSI) is a research hotspot in the field of remote sensing. It is of particular importance in many domains, especially in military application. Unsupervised target detection is usually more difficult because there is no prior information about target. Traditional algorithms exploit spectral information, only. This study introduces the idea of saliency detection from the visual technique into HSI processing domain and proposes a novel approach named spectral saliency target detection (SSD). It establishes a novel salient model, which utilises both spatial saliency and spectral saliency. In the framework of SSD, it combines the model with spectral matching algorithm to make it perform well even in situations where the target is concealed and small. A HSI set comprised of eight different scenes with complex background is setup to evaluate the performance of the proposed algorithm. The final visible detection results demonstrate that the SSD algorithm outperforms the others. The receiver operation characteristic (ROC) curve and area under the ROC curve are applied to evaluate the results. The proposed algorithm shows superior and stable performance.