The application of diffusion models in the field of remote sensing image generation has significantly improved the performance of generation algorithms. However, existing methods still exhibit certain limitations, such as the inability to generate images with rich texture details and minimal geometric distortions in a controllable manner. To address these shortcomings, this work introduces an innovative remote sensing image generation algorithm, OP-Gen, which is guided by textual descriptions and OpenStreetMap (OSM) images. OP-Gen incorporates two information extraction branches: ControlNet and OSM-prompt (OP). The ControlNet branch extracts structural and spatial information from OSM images and injects this information into the diffusion model, providing guidance for the overall structural framework of the generated images. In the OP branch, we design an OP-Controller module, which extracts detailed semantic information from textual prompts based on the structural information of the OSM image. This information is subsequently injected into the diffusion model, enriching the generated images with fine-grained details, aligning the generated details with the structural framework, and thus significantly enhancing the realism of the output. The proposed OP-Gen algorithm achieves state-of-the-art performance in both qualitative and quantitative evaluations. The qualitative results demonstrate that OP-Gen outperforms existing methods in terms of structural coherence and texture detail richness. Quantitatively, the algorithm achieves a Fréchet inception distance (FID) of 45.01, a structural similarity index measure (SSIM) of 0.1904, and a Contrastive Language-Image Pretraining (CLIP) score of 0.3071, all of which represent the best performance among the current algorithms of the same type.

Aiming to address the fundamental limitation of fixed activation functions that constrain network expressiveness in existing deep deblurring models, in this pioneering study, we introduced Kolmogorov–Arnold Networks (KANs) into the field of full-color/RGB image deblurring, proposing the Kolmogorov–Arnold and Transformer Network (KT-Deblur) framework based on dynamically learnable activation functions. This framework overcomes the constraints of traditional networks’ fixed nonlinear transformations by employing adaptive activation regulation for different blur types through KANs’ differentiable basis functions. Integrated with a U-Net architecture within a generative adversarial network framework, it significantly enhances detail restoration capabilities in complex scenarios. The innovatively designed Unified Attention Feature Extraction (UAFE) module combines neighborhood self-attention with linear self-attention mechanisms, achieving synergistic optimization of noise suppression and detail enhancement through adaptive feature space weighting. Supported by the Fast Spatial Feature Module (FSFM), it effectively improves the model’s ability to handle complex blur patterns. Our experimental results demonstrate that the proposed method outperforms existing algorithms in terms of peak signal-to-noise ratio (PSNR) and structural similarity (SSIM) metrics across multiple standard datasets, achieving an average PSNR of 41.25 dB on the RealBlur-R dataset, surpassing the latest state-of-the-art (SOTA) algorithms. This model exhibits strong robustness, providing a new paradigm for image-deblurring network design.

Single-lens imaging systems can capture spectral information, but they are affected by aberrations, defocusing, and other factors, leading to spectral data overlap. It is then necessary to perform restoration on overlapped spectral data. The restoration method should not only deblur image information, but also recover spectral information for each wavelength band. In other words, the restoration process needs to handle both spatial and spectral information in parallel while ensuring that neither is distorted. In this study, while considering the characteristics of capturing overlapped information in a single-lens imaging system, a spectrum restoration method based on gradient prior information optimization is proposed. This method is shown to achieve high-quality restoration of both images and spectra. The feasibility of this algorithm is demonstrated through simulation and experimental verification, which show that the restoration quality of spectra using the proposed algorithm is improved compared to that achieved using the MTV algorithm.

Scale variation presents a significant challenge in object detection. To address this, multi-level feature fusion techniques have been proposed, exemplified by methods such as the feature pyramid network (FPN) and its extensions. Nonetheless, the input features provided to these methods and the interaction among features across different levels are limited and inflexible. In order to fully leverage the features of multi-scale objects and amplify feature interaction and representation, we introduce a novel and efficient framework known as a multi-resolution and semantic-aware bidirectional adapter (MSBA). Specifically, MSBA comprises three successive components: multi-resolution cascaded fusion (MCF), a semantic-aware refinement transformer (SRT), and bidirectional fine-grained interaction (BFI). MCF adaptively extracts multi-level features to enable cascaded fusion. Subsequently, SRT enriches the long-range semantic information within high-level features. Following this, BFI facilitates ample fine-grained interaction via bidirectional guidance. Benefiting from the coarse-to-fine process, we can acquire robust multi-scale representations for a variety of objects. Each component can be individually integrated into different backbone architectures. Experimental results substantiate the superiority of our approach and validate the efficacy of each proposed module.