Adaptive ability is essential for a quadrotor to fly in a complex environment with various situations thus a lightweight motion planner is necessary to ensure the quadrotor’s maneuverability. At present, the ESDF-free framework has become a mainstream technology in the field of local motion planning. However, it models the collision cost explicitly which increases computational burden. In this article, an adaptive motion planning method is proposed based on decision making within the ESDF-free framework for quadrotors. At first, a decision making model is constructed based on the finite-state machine(FSM) which can improve the adaptive ability of quadrotors in complex environment. Then a local trajectory can be obtained within the ESDF-free framework and collision cost is considered implicitly through fitting a safe trajectory. The proposed method has been validated with a large number of experiments and it has shown excellent performance compared with the state-of-the-art methods.

Existing Infrared and Visible Image Fusion (IVIF) methods typically assume high-quality inputs. However, when handing degraded images, these methods heavily rely on manually switching between different pre-processing techniques. This decoupling of degradation handling and image fusion leads to significant performance degradation. In this paper, we propose a novel VLM-Guided Degradation-Coupled Fusion network (VGDCFusion), which tightly couples degradation modeling with the fusion process and leverages vision-language models (VLMs) for degradation-aware perception and guided suppression. Specifically, the proposed Specific-Prompt Degradation-Coupled Extractor (SPDCE) enables modality-specific degradation awareness and establishes a joint modeling of degradation suppression and intra-modal feature extraction. In parallel, the Joint-Prompt Degradation-Coupled Fusion (JPDCF) facilitates cross-modal degradation perception and couples residual degradation filtering with complementary cross-modal feature fusion. Extensive experimental results indicate that the proposed VGDCFusion demonstrates marked superiority in degraded image fusion tasks, surpassing existing state-of-the-art methods in both qualitative visual quality and quantitative evaluation metrics (e.g., the AG and SF measures achieve average improvements of approximately 15% and 14.75%, respectively). Our code is available at https://github.com/Lmmh058/VGDCFusion.

Fourier ptychographic microscopy (FPM) enables wide-field, high-resolution complex imaging, but large defocus caused by sample unevenness, positioning errors, and system instability can severely degrade reconstruction quality. Under large-defocus conditions, the sample spectrum and pupil phase are strongly coupled, making pupil recovery sensitive to noise and nonphysical phase fluctuations. We propose a Zernike-ADMM framework for Fourier ptychographic microscopy (ZA-FPM), which imposes physics-constrained pupil regularization for large-defocus reconstruction. The key idea is to reformulate the unstable pupil update under large defocus as a physics-constrained wavefront estimation problem. Specifically, the pupil amplitude is fixed by the objective aperture support, while the pupil phase is represented by low-order Zernike wavefront modes. To implement the Zernike phase representation, a lightweight Zernike-fitting neural network (ZFNN) is introduced into the alternating direction method of multipliers (ADMM) reconstruction process and optimized online to generate the corresponding Zernike coefficients. This training-free design avoids offline pretraining, labeled defocus distances, and sample-specific datasets. Experiments demonstrate that ZA-FPM reliably recovers major sample structures and maintains a stable pupil phase even under strong noise with a 150um defocus, offering an effective solution for FPM reconstruction under non-ideal conditions.

Speckle correlation imaging based on the optical memory effect provides a valuable approach for non-invasive imaging through scattering media, yet its practical implementation faces significant challenges in reconstruction speed. To overcome these challenges, a fast speckle correlation imaging (FSCI) algorithm that incorporates three key technical innovations is presented in this paper. The method commences with the intelligent sub-speckle quality control and initialization adaptation (ISQIA) module, which integrates two core functional sub-modules: one is sub-speckle screening (SS), which eliminates low-information sub-speckles based on the entropy-contrast criteria; the other is adaptive initialization (AI), which selects optimal sub-speckles to accelerate the convergence of subsequent calculations. The reconstruction accuracy is further enhanced through weighted coherent averaging (WCA), which utilizes cross-correlation peaks as weighting coefficients. Finally, dynamic iteration termination (DIT) automatically halts computation when reconstruction changes become negligible, thus optimizing computational efficiency. Experimental results show that the FSCI algorithm is 40.9% faster than traditional speckle correlation imaging methods, taking only 3.13 s for reconstruction in darkrooms. It also improves imaging quality significantly: PSNR reaches 27.04 dB, and SSIM is improved to 0.87. Notably, FSCI performs stably in weak to extremely strong noise (SNR 1.79 to -5.78dB), excels at recovering complex structures like Chinese characters, and thus has high application potential in fields such as biomedical imaging and industrial non-destructive testing.

Most existing infrared-visible image fusion (IVIF) methods assume high-quality inputs, and therefore struggle to handle dual-source degraded scenarios, typically requiring manual selection and sequential application of multiple pre-enhancement steps. This decoupled pre-enhancement-to-fusion pipeline inevitably leads to error accumulation and performance degradation. To overcome these limitations, we propose Guided Dual-Domain Fusion (GD^2Fusion), a novel framework that synergistically integrates vision-language models (VLMs) for degradation perception with dual-domain (frequency/spatial) joint optimization. Concretely, the designed Guided Frequency Modality-Specific Extraction (GFMSE) module performs frequency-domain degradation perception and suppression and discriminatively extracts fusion-relevant sub-band features. Meanwhile, the Guided Spatial Modality-Aggregated Fusion (GSMAF) module carries out cross-modal degradation filtering and adaptive multi-source feature aggregation in the spatial domain to enhance modality complementarity and structural consistency. Extensive qualitative and quantitative experiments demonstrate that GD^2Fusion achieves superior fusion performance compared with existing algorithms and strategies in dual-source degraded scenarios. The code will be publicly released after acceptance of this paper.

Speckle correlation-based scattering imaging excels in rapid imaging due to its compact setup and computational reconstruction capability. However, most existing methods for improving the performance of speckle correlation imaging (SCI) not only over-rely on enhancement coefficient tuning but also lack robustness in terms of fusion coefficients. To address this issue, we propose a method for speckle imaging that introduces an adaptive enhancement scheme. This framework integrates multi-scale morphological gradients with wavelet-domain cross-scale autocorrelation to construct a Multi-scale Edge Significance Factor (MESF), enabling spatially-adaptive optimization of enhancement coefficients through high-frequency detail emphasis. Experiments confirm that under complex lighting, specifically wide-band light and non-darkroom conditions, SC-MSUM significantly improves speckle contrast and reconstruction quality. Its modular design allows it to serve as a preprocessing step or for post-reconstruction edge enhancement within various SCI workflows, demonstrating promising application potential.

The infrared and visible images fusion (IVIF) is receiving increasing attention from both the research community and industry due to its excellent results in downstream applications. However, existing deep learning methods exhibit limitations in global feature modeling, balancing fusion performance with computational efficiency and effectively leveraging frequency-domain information. To address this limitation, we propose an end-to-end fusion network named the Frequency-Spatial Attention Transformer Fusion Network (FSATFusion). The FSATFusion contains the frequency-spatial attention Transformer (FSAT) module designed to effectively capture discriminate features from source images. The FSAT module includes a frequency-spatial attention mechanism (FSAM) capable of extracting significant features from feature maps. Additionally, we propose an improved Transformer module (ITM) to enhance the ability to extract global context information of vanilla Transformer without incurring additional computational overhead. Across four public datasets (TNO, MSRS, RoadScene, and RGB-NIR), we conducted extensive qualitative comparisons and quantitative evaluations based on eight metrics against fourteen representative state-of-the-art fusion algorithms. Experimental results demonstrate that the proposed method outperforms state-of-the-art deep learning approaches (e.g., GANMcC, MDA, and EMMA) in terms of qualitative visual quality, objective metrics (e.g., achieving an average improvement of approximately 34% in MI, 5% in Qy, and 4% in VIF), as well as computational efficiency. Furthermore, the fused images generated by our method exhibit superior applicability and performance in downstream object detection tasks. Our code is available at https://github.com/Lmmh058/FSATFusion.

Infrared small target detection (IRSTD) is widely recognized as a challenging task due to the inherent limita tions of infrared imaging, including low signal-to-noise ratios, lack of texture details, and complex background interference. While most existing methods model IRSTD as a semantic segmentation task, but they suffer from two critical drawbacks: (1) blurred target boundaries caused by long-distance imaging dispersion; and (2) ex cessive computational overhead due to indiscriminate feature stacking. To address these issues, we propose the Lightweight Efficient Infrared Small Target Detection (LE-IRSTD), a lightweight and efficient framework based on YOLOv8-n, featuring the following key innovations. Firstly, we identify that the multiple bottleneck structures within the C2f component of the YOLOv8-n backbone contribute to an increased computational bur den. Therefore, we implement the Mobile Inverted Bottleneck Convolution block (MBConvblock) and Bottleneck Structure block (BSblock) in the backbone, effectively balancing the trade-off between computational efficiency and the extraction of deep semantic information deep semantic feature extraction. Secondly, we introduce the Attention-based Variable Convolution Stem (AVCStem) structure, substituting the final convolution with Variable Kernel Convolution (VKConv), which allows for enables adaptive convolutional kernels that can transform into various shapes, facilitating the receptive field for the extractionof targets thereby facilitating receptive field ex pansion for small-target feature extraction. Finally, we employ Global Shuffle Convolution (GSConv) to shuffle the channel dimension channel-dimension features obtained from different convolutional approaches, thereby en hancing the robustness and generalization capabilities of our method. Experimental results demonstrate that our LE-IRSTD method achieves compelling results achieves compelling performance in both accuracy and lightweight performance efficiency, outperforming several state-of-the-art deep learning methods. The source codes are available at https://github.com/jing2024star/LE-IRSTD.

Optical image-based texture and material surface recognition is important for computational imaging, intelligent inspection, and material-reference retrieval, where a system often needs not only a closed-set label but also inspectable evidence and an expandable gallery. Scale-varying micro-textures, repeated surface structures, global layout cues, and visually ambiguous categories make this task difficult under non-controlled imaging conditions. This paper presents Reliability-Aware Global-Local Metric Retrieval (RAGL-MR), which maps images and patches into a shared ArcFace metric space, retrieves multi-scale patch neighbors as local surface evidence, compresses whole-image reference galleries using Multi-Prototype Global Distribution Modeling (MP-GDM), and combines branch scores through validation-selected reliability-aware fusion. The framework is evaluated on public texture/material surface image benchmarks used as proxies for surface image analysis rather than as controlled photonic experiments. On the first five official DTD splits, the final fusion reaches 74.06 ± 0.11 Top-1, 92.61 ± 0.61 Top-5, and 96.11 ± 0.23 Top-10; on MINC-2500, it reaches 85.84 ± 0.25 Top-1, 97.63 ± 0.28 Top-5, and 99.15 ± 0.10 Top-10, with additional checks on FMD and KTH-TIPS2-b. Additional perturbation, calibration, large-gallery, and industrial surface-defect checks further examine robustness, reliability, and practical retrieval behavior. These results indicate that RAGL-MR provides an evidence-aware and gallery-extensible retrieval complement for optical material surface image analysis.

Recovering high-quality images from low-quality speckle patterns remains a core challenge in scattering imaging, especially under narrowband illumination with ambient light interference and broadband illumination. This paper proposes a dual-input synchronous transmission architecture: after Correction-Smoothing-Phase Optimization (CSPO) preprocessing, two data streams are parallel-fed into Phase-Aligned Coherent Summation (PACS) for efficient and high-precision reconstruction with adaptive fusion, breaking the single-path limitation of traditional methods and balancing imaging efficiency and quality. Additionally, an adaptive enhancement factor feedback mechanism is designed for Median-Unsharp Sharpening Enhancement (MUSE) to dynamically adjust Median Filtering (MF) and Unsharp Masking (USM) parameters, achieving adaptive balance between noise suppression and detail enhancement and improving robustness under extreme lighting. In PACS, a dynamic reference update mechanism is introduced, combined with fixed amplitude to realize iterative phase optimization, effectively suppressing speckle noise and boosting the signal-to-noise ratio of reconstructed images. Experimental results show that the proposed method achieves favorable restoration performance even at a SNR of −8.7 dB under narrowband and broadband illumination with spectral bandwidths of 100 nm, 200 nm, and 280 nm (FWHM), and significantly improves image quality in unknown scattering media, showing great potential for robust speckle reconstruction.

While image deblurring has made significant progress recently, existing methods still suffer from a challenge in handling complex and diverse blur images caused by camera shake and object movement. Inspired by the observation that different images are degraded in dissimilar ways, we propose to explore image-level degradation knowledge to perform deblurring for complex blur images. Consequently, we propose a novel framework for image deblurring, termed MDPL, which leverages degradation-aware visual priors to enhance performance on complex and diverse blurry images. First, we introduce a Momentum Contrast Feature Module (MCFM) that explicitly models image-level degradation-aware knowledge as visual priors, enriching features with blur-related cues. Second, we design a Blur-Adaptive Relational Fusion (BARF) module, which embeds these priors into the hierarchical features of a Transformer-enhanced architecture and adaptively captures feature interactions under varying blur patterns. Benefiting from the modeling of rich degradation knowledge, MDPL effectively enhances deblurring performance on complex and diverse blur images. Extensive experiments demonstrate that MDPL achieves state-of-the-art performance. Code is available at https://github. com/Maykkkkk/MDPL.

Optical clearing agents (OCAs) can effectively reduce tissue scattering and improve photon transmission, facilitating non-invasive optical diagnostics. To quantitatively describe their optical clearing effects in multilayered skin, we developed a hybrid modeling framework that combines a multilayer Monte Carlo (MC) simulation with a diffusion model governed by Fick's law. The MC component simulates photon transport through stratified skin layers with varying optical properties, while the diffusion model characterizes the temporal evolution of OCA concentration and refractive index. By integrating these two processes, the model enables depth- and time-resolved analysis of light propagation under dynamic optical clearing conditions. This approach provides a theoretical basis for evaluating OCA performance across different skin regions and supports optimization of parameters for improved optical penetration in diagnostic applications.

Deep learning techniques have revolutionized the infrared and visible image fusion (IVIF), showing remarkable efficacy on complex scenarios. However, current methods do not fully combine frequency domain features with global semantic information, which will result in suboptimal extraction of global features across modalities and insufficient preservation of local texture details. To address these issues, we propose Wavelet-Mamba (W-Mamba), which integrates wavelet transform with the state-space model (SSM). Specifically, we introduce Wavelet-SSM module, which incorporates wavelet-based frequency domain feature extraction and global information extraction through SSM, thereby effectively capturing both global and local features. Additionally, we propose a cross-modal feature attention modulation, which facilitates efficient interaction and fusion between different modalities. The experimental results indicate that our method achieves both visually compelling results and superior performance compared to current state-of-the-art methods. Our code is available at https://github.com/Lmmh058/W-Mamba.

Magnetic resonance imaging (MRI) is an advanced imaging technique that is used to aid in medical diagnosis. However, common noises such as Gaussian and Rician noise can blur details and structures, affect contrast and reduce signal-to-noise ratio (SNR), so MRI denoising technique becomes an critical step to get noise-free MRI images. Traditional methods still have limitations in effectively balancing noise removal and the preservation of image details and structural information. To address the challenge, this paper proposes an MRI image denoising model that combines Nonlinear Mapping Network (NLMap) and Attention Mechanism-guided Adaptive Total Variation Regularization (ATVR). The model includes a NLMap-ATVR network, a crafted joint loss function and a Bayesian optimization framework. Firstly, the network uses an encoder-decoder architecture, combined with ATVR to ensure noise removal. Secondly, the joint loss function includes mean square error (MSE) loss, perceptual loss and ATVR loss, which are used to consider pixel-level and feature-level spatial structural errors to preserve details and structures. Thirdly, a Bayesian optimization framework is applied to automatically tune the hyperparameters to obtain optimal parameters. Compared with State-of-the-art methods, both subjective and objective evaluations based on experimental results demonstrate that the proposed method not only effectively removes noise but also significantly preserves details and structural information, which greatly improves SNR.

Medical image fusion is crucial in clinical applications, combining data from various medical imaging modalities into a single high-quality image to enhance diagnosis. However, existing fusion algorithms suffer from several limitations, including inadequate feature extraction, leading to detail loss, poor inheritance of complementary information between modalities, and insufficient evaluation of color information in color and grayscale fusion tasks. To address these challenges, we propose a novel, to our knowledge, medical image fusion framework based on the denoising diffusion probabilistic model. Our model adopts a two-stage training strategy: feature extraction and image reconstruction. An edge-enhancement dense block is designed to work with a denoising diffusion probabilistic model as a feature extractor, learning and extracting joint features from multimodal medical images to ensure comprehensive feature extraction. To further integrate meaningful information and enhance the visual quality of fused images, we design a feature-enhanced reconstruction network that amplifies features during the reconstruction process. Additionally, we develop distinct joint loss functions based on the strengths and weaknesses of different modalities, ensuring effective retention of complementary information. In the color and grayscale fusion task, we introduce a multi-channel joint learning method to ensure the retention of complementary information and incorporate a color difference formula to evaluate color retention. Experimental results demonstrate that our proposed method significantly outperforms existing state-of-the-art techniques, producing fused images with improved clarity, enhanced detail preservation, and more effective inheritance of complementary information across modalities.

Speckle correlation-based scattering imaging is an ingenious field, as it allows for the efficient reconstruction of object images using computational techniques in a simple setup. However, this method typically necessitates high-contrast speckle images captured in a darkroom environment, restricting its applicability to specific scenarios. Here, we present a fast and non-darkroom imaging framework, namely FNDI, for reconstructing objects through scattering media under ambient light interference. Specifically, a speckle illumination model is established guided by the total variational Retinex (TV-Retinex) theory, and the speckle illumination component is adjusted to obtain an enhanced speckle with significantly improved contrast. Then, a modified Fienup algorithm with the iteration-driven shrinkwrap (IDS) strategy is employed to rapidly reconstruct the object image through tens of iterations. Extensive experiments are conducted under different lighting conditions to evaluate FNDI in comparison with existing non-darkroom methods and the classical speckle correlation method. The results demonstrate that FNDI is effective and efficient, making it highly attractive for practical scattering imaging applications.

The dehazing algorithm based on the dark channel prior theory is a widely used dehazing method. Despite successfully enhancing clarity in foggy images, it still has some limitations such as its effectiveness diminishes when images turn grayscale, indicating an indistinct dark channel in the scene. Based on the fundamentals of the dark channel prior dehazing algorithm, an improved dehaze algorithm called LINTI-DCP, tailored for laser illumination nighttime images was developed. LINTI-DCP can overcome the inaccurate estimation of atmospheric light and transmission rates in foggy infrared laser night vision. Experimental results demonstrated that LINTI-DCP is an effective dehazing algorithm for laser illumination nighttime images.

Despite the complex optical lens system, classical optical microscopy cannot achieve high spatial resolution and a large field of view (FOV). The lensless optical ptychographic microscopy technique solves the contradiction between resolution and FOV with a simple optical path. However, position errors between LED array and sample, caused by misalignment of optical imaging axis and manufacture errors of LED array, decrease reconstruction quality. Due to the difficulty of hardware correction, we propose an adaptive genetic strategy to correct LED array position error for lensless ptychographic imaging. Firstly, the coarse positions of the LEDs are predicted by Kalman filter, thus we can determine the position range. Secondly, the precise positions of LEDs are calculated by our adaptive genetic algorithm, which combines the fitness function according to the previous predicted positions. Finally, we incorporate the LED array position correction into the Gerchberg-Saxton (GS) phase recovery method to achieve improved reconstructed image quality. Compared with similar methods, both simulation and real experiments indicate that our method genetic algorithm adapt-GS (adaptive Genetic algorithm-based GS) method has higher robustness in position errors correction, and creates results with better quality. This method is applied to other state-of-art reconstruction algorithms as a pre-processing step, highly improve the image quality.