Search Results (844)

Object detection in complex industrial environments is prone to being affected by insufficient dynamic weighting of local and global features, as well as illumination variations and impurities. Moreover, existing models suffer from excessive model complexity, which directly impairs computational efficiency. To more accurately distinguish Chinese herbal materials with diverse morphologies, this paper proposes the MobileAttn module. Drawing on the idea of token representation in the Transformer architecture, this module extracts contextual information through global feature compression, fuses it with tokens to generate a spatial attention map, and realizes dynamic recalibration of convolutional features. This process enhances the feature weights of key semantic regions, suppresses redundant background information, and improves feature discriminability. To address illumination interference, brightness-aware weights are combined with dual-path (channel and spatial) attention for global control, dynamically reducing the impact of illumination; this component is named LightAttn. When Chinese herbal materials contain common industrial unknown impurities (e.g., small stones and weeds), an impurity detection auxiliary module, a post-processing step independent of the main detection network, is proposed. This module refines Non-Maximum Suppression (NMS) logic to distinguish target Chinese herbal materials from interfering impurities. Subsequently, it accurately locates and marks impurities on the conveyor belt, thereby achieving effective unknown impurity detection. Experimental results demonstrate that, compared with the original YOLOv11 on the Chinese herbal materials detection task, the optimized model achieves a 1.7% improvement in the overall mean Average Precision (mAP@0.5:0.95). On a per-class basis, gains are particularly pronounced for certain challenging high-aspect-ratio Chinese herbal materials. Prunella vulgaris and orange peel achieve respective AP improvements of 5.8% and 4.1%. Meanwhile, the model parameter count is reduced by 23.1% and the computational complexity by 20.3%. The F1-Score of the impurity detection results is 86.38%, verifying the effectiveness of the impurity detection auxiliary module. Full article

Object detection in low-light scenes is difficult because weak illumination reduces local contrast, amplifies sensor noise, and makes small or occluded objects hard to localize. Existing enhancement-before-detection pipelines can improve visual brightness, but they are not always optimized for detection features, while transformer-style global reasoning is often too costly for lightweight detectors. To address this gap, we propose GE-Detection, a detector-side framework that integrates Global Sub-Sampled Attention (GSA), Efficient Multi-scale Attention (EMA), and dropout regularization into YOLO- and PicoDet-style architectures. GSA introduces lower-cost global context modeling through spatially reduced key-value tokens, EMA refines multi-scale fused features without aggressive channel compression, and dropout improves training-time regularization with no inference-time parameter overhead. Experiments on COCO, ExDark, BDD100K-Night, and NightOwls show that the method is most effective in low-light detection: on ExDark with YOLO11n, mAP50-95 improves from 34.39% to 36.74%, mAP50 from 56.24% to 59.27%, and Box (P) from 67.63% to 71.36%. The full YOLO11n variant uses 2.91M parameters and maintains 134.7 FPS on an RTX 2080 Ti under the tested setting. Cross-dataset and corruption experiments further indicate that the proposed modules improve localization under several nighttime domain shifts while retaining known limitations under severe noise and adverse weather. These results indicate that combining efficient global attention, multi-scale feature recalibration, and targeted regularization can improve low-light localization while keeping the detector practical for deployment. Full article

Observation of welding features is important for GMAW training and instruction because the welding arc, molten pool, filler wire, and groove can be difficult to distinguish during welding. In this study, a compact, low-cost, and practical visualization system was developed to support gas metal arc welding (GMAW) skill training from both the welder’s and instructor’s perspectives. The system consists of a welder-side unit and an instructor-side unit and uses a commercial camera, optical filters, a wide-angle lens, and a compact computer. Welding images were acquired under actual GMAW conditions, and the effects of optical filter selection, exposure time, tone mapping, and trimming methods were investigated. A 600 nm long-pass filter and an exposure time of 20,000 μs provided a suitable balance between arc-light suppression, brightness stability, and image clarity. Gamma correction improved the visibility of key regions, including the molten pool, arc, torch, groove, and wire. In addition, low-pass-filtered centroid tracking enabled stable trimming of the weld region from wide-angle images. The developed system achieved real-time display and recording of standardized welding images, demonstrating its potential to support GMAW training through improved image visibility, real-time monitoring, and standardized image recording, while also providing visual data for post-weld review and future skill-assessment applications. Full article

Organic–inorganic hybrid metal halides (OIMHs), especially zero-dimensional (0D) ones, have been recognized as an excellent class of luminescent materials due to their structural diversity and tunable emission properties. In this work, using the environmentally friendly Zn(II) ion as the central metal and 1,4,7,10-tetraazacyclododecane (Cyclen) as the organic component, we successfully synthesized a novel OIMH, (H₃Cyclen)(ZnBr₄)·Br·H₂O. Single-crystal X-ray diffraction analysis reveals that (H₃Cyclen)(ZnBr₄)·Br·H₂O possesses a 0D structure, in which the [ZnBr₄]²⁻ tetrahedra are uniformly separated by the organic amine cations. This structural feature is expected to enhance the material’s stability and optimize its optoelectronic properties. Under UV lamp irradiation, (H₃Cyclen)(ZnBr₄)·Br·H₂O emits bright blue-green light. Therefore, we systematically investigated its luminescence properties. The emission mechanism was further elucidated using UV–vis absorption spectroscopy and DFT calculations. Finally, (H₃Cyclen)(ZnBr₄)·Br·H₂O was employed as a luminescent material to fabricate a white light-emitting diode (WLED), demonstrating its potential as an excellent phosphor material. Full article

This study evaluates the potential of L-band passive microwave data for monitoring soil moisture (SM) in boreal and temperate forests using SMAP and SMOS AM and PM overpasses. SMAP and SMOS Level 3 SM products were first assessed for spring and summer seasons. SMOS showed lower accuracy (r² = 0.04–0.24, ubRMSE = 0.09–0.13 m³/m³), while SMAP performed better (r² = 0.18–0.62, ubRMSE = 0.05–0.07 m³/m³) across sites and overpasses. Given the larger number of SMAP TB observations at a fixed incidence angle and greater temporal coverage over the study area, SMAP was selected for SM estimation using ML models. Feature importance analysis identified brightness temperature (TB) as the most influential variable, followed by vegetation water content (VWC), air and soil temperatures, and the microwave polarization difference index (MPDI). Soil and air temperatures were interchangeable during AM overpasses, whereas PM overpasses showed distinct differences, likely due to thermal absorption by dense vegetation. Using optimal features, SM was estimated with CatBoost, Gradient Boosting (GB), Random Forest (RF), and Principal Component Regression (PCR), using stratified shuffle split (SSS) and leave-one-year-out cross-validation (LOYOCV). In SSS, CatBoost achieved slightly higher accuracy than the other ensemble models (AM: r² = 0.73; PM: R² = 0.74), while PCR yielded substantially lower accuracy across both overpasses. LOYOCV showed closer rankings among models, with CatBoost ranking highest overall (r² = 0.58 for AM and 0.54 for PM). Results highlight the feasibility of improved SM estimation in forests using L-band TB, VWC, temperature variables, and MPDI. Full article

High dynamic range (HDR) image reconstruction from a single low dynamic range (LDR) input remains an important problem for computational photography, particularly when practical deployment on consumer-grade hardware is considered. With the increasing availability of hardware supporting HDR, public demand for capturing and viewing HDR images has grown significantly. Recent research has explored deep learning-based approaches to reconstruct HDR images from low dynamic range (LDR) inputs by extracting regional pixel features or leveraging the camera response function (CRF) for model training. Many of these approaches employ Convolutional Neural Network (CNN) architectures and utilize skip connections to preserve learned information. Nevertheless, the configuration-level effects of data augmentation in HDR reconstruction remain insufficiently discussed. Existing CNN-based approaches, such as HDRCNN, HDRUNet, and ExpandNet, have demonstrated promising reconstruction ability, but they may involve a heavy backbone architecture, a long training time, or a limited discussion of how preprocessing configurations affect reconstruction performance. This study presents an engineering-oriented HDR reconstruction framework derived from HDRCNN, focusing on practical efficiency, structural fidelity, and training feasibility. The proposed framework introduces three modifications: (1) a configuration-level comparison of composite data augmentation settings, including unsharp masking, denoising, Gaussian blur, and brightness–contrast adjustment; (2) the replacement of the original VGG16 backbone with a ResNet50-based encoder enhanced with attention blocks and squeeze-and-excitation (SE) blocks for improved multi-scale feature extraction and channel-wise recalibration; and (3) the integration of mixed-precision training with cosine annealing learning-rate scheduling to reduce computational cost. Experimental results on the SI-HDR dataset show that the best composite augmentation configuration improves PSNR from 19.05 dB to 22.10 dB and SSIM from 0.6444 to 0.7714 without increasing the training time. Compared with the original VGG16-based HDRCNN setting, the ResNet50-based model reduces training time while improving SSIM from 0.2705 to 0.8512. Under the adopted comparison protocol, the proposed model achieves the shortest training time and slightly higher PSNR than HDRUNet, while HDRUNet retains a higher SSIM. This indicates a trade-off among pixel-wise fidelity, structural similarity, and computational efficiency. The current evaluation is limited by a small test setting, composite rather than operation-level augmentation analysis, and the use of PSNR and SSIM only; therefore, future work should include full benchmark evaluation, additional perceptual/HDR-specific metrics, and controlled component-level ablation studies. Full article

Blind image watermarking models such as HiDDeN have laid an important foundation for end-to-end watermarking. Nevertheless, they still suffer from three major limitations: single-scale feature extraction, fixed fusion weights, and slow training convergence. To address these issues, this paper proposes an adaptive multi-scale watermarking algorithm based on collaborative dual-attention mechanisms. The algorithm designs an adaptive multi-scale feature fusion module (MA-FFM) with a dynamic gating network in the encoder, which flexibly combines local multi-scale textures with global contextual information, overcoming the limitation of fixed fusion weights. In the decoder, a multi-level channel attention module is embedded to strengthen the extraction of watermark signals. The two attention modules work synergistically: the encoder focuses on adaptive feature fusion while the decoder leverages channel attention to selectively enhance watermark-related features, forming a dual-attention synergy that balances robustness and imperceptibility. Moreover, the dynamic gating network adaptively adjusts the contribution of local versus global features via learnable weights, whose evolution from approximately 0.51 to about 0.89 improves model interpretability. Experiments are conducted on the COCO 2017 dataset. Compared with HiDDeN, the proposed algorithm reduces the bit error rate (BER) from 0.1696 to 0.1538 under no attack with a relative reduction of 9.3%, increases PSNR by 0.61 dB, and improves SSIM from 0.9058 to 0.9077. Under various attacks—including JPEG compression, Gaussian noise, salt-and-pepper noise, and brightness/contrast adjustments—the BER remains consistently lower than that of HiDDeN. Ablation studies confirm the effectiveness of each module. Overall, the proposed algorithm preserves visual quality, improves the accuracy of watermark embedding and extraction, and exhibits strong generalization robustness against common image distortions. Full article

Microwave-based remote sensing possesses the capability to penetrate through atmospheric obstructions such as cloud layers and fog, making it extensively utilized for estimating parameters including soil water content, atmospheric moisture levels, and terrestrial surface temperatures. Extended temporal datasets serve as fundamental requirements for climatological investigations; however, individual satellite operational lifespans remain constrained and prove inadequate for establishing multi-decade temporal sequences. Consequently, conducting comparative analyses and implementing cross-calibration procedures across measurements obtained from distinct sensors exhibiting comparable operational features becomes imperative. The FengYun (FY)-3G spacecraft, deployed into orbit during April 2023, hosts China’s most recent orbiting microwave radiometric instrument, designated as the Microwave Radiation Imager–Rainfall Mission (MWRI-RM). The FY-3G satellite’s unique drifting equator crossing time orbit plays a critical role in the calibration behavior of the MWRI-RM instrument, representing a key novelty of this study. The reliability of its brightness temperature (TB) observations has attracted considerable attention. Within this investigation, we conduct comparative assessments of orbital TB observations acquired from FY-3G/MWRI-RM against corresponding measurements obtained from the Advanced Microwave Scanning Radiometer 2 (AMSR2) installed on the Global Change Observation Mission–Water 1 (GCOM-W1) platform, and establish a straightforward linear inter-calibration methodology. Both sensing systems show strong consistency, with correlation coefficients exceeding 0.9 for all corresponding channels and systematic biases ranging from −1.40 K to −0.14 K. FY-3G/MWRI-RM generally reports lower TB values than GCOM-W1/AMSR2. The inter-sensor differences vary with frequency, land cover type, and TB range. Larger negative biases are mainly observed at 23.8 GHz and over water bodies, whereas the biases at 89 GHz are generally close to zero for most surface types. Latitude-dependent TB biases are most evident at 10.65 and 18.7 GHz, especially for vertical polarization at high latitudes, while orbit-dependent differences are more pronounced for vertically polarized low- and mid-frequency channels. After applying an inter-calibration procedure using AMSR2 as the reference, the agreement between FY-3G/MWRI-RM and GCOM-W1/AMSR2 is improved substantially, with mean biases below 0.25 K and RMSE values below 2 K for all channels. Validation using independent datasets further supports the stability of the calibration. The calibrated FY-3G/MWRI-RM TB data provide a basis for constructing long-term passive microwave brightness temperature records and for retrieving land and atmospheric parameters. Full article

The Event Horizon Telescope (EHT) imaging of M87* and Sgr A* provides a unique opportunity to test spacetime geometries in the strong-field regime. Motivated by this, we systematically investigate the optical characteristics for three types of nonsingular black holes (BHs) with a Minkowski core and constrain the quantum gravity effect parameter $\alpha$ and the regularization parameter n using EHT observational data. Utilizing the observed shadow sizes of M87* and Sgr A*, we conduct a detailed comparison of the constraints on $\alpha$ for the three BH types. Our analysis reveals significant differences among them: Type I BHs exhibit the largest upper limit, whereas Type III BHs show the smallest upper limit. Furthermore, the constraints derived from M87* observations are tighter than those from Sgr A*, reflecting the close dependence of these limits on current observational precision. Subsequently, we simulate BH images at the current EHT resolution using a Gaussian filter. Although the photon ring and lensed ring features cannot be resolved, variations in shadow size and brightness distribution are clearly detectable. Within the parameter space allowed by EHT observations, the shadow size and total intensity exhibit a distinct monotonic hierarchy: Type I BHs display the largest shadow and highest total intensity, while Type III BHs show the opposite trend. Finally, we find that increasing $\alpha$ leads to shadow contraction and dimming, whereas increasing n causes the shadow to expand while making the optical characteristics of the three BH types increasingly indistinguishable. Consequently, the three BH types become more readily distinguishable only when n is small or $\alpha$ is large. Full article

Affected by global climate change and complex environmental factors, the frequency and intensity of forest fires have been rising. Accurate early detection is crucial for disaster mitigation. Traditional methods (e.g., manual monitoring) suffer from low efficiency or limited coverage, while deep learning methods (e.g., YOLO (You Only Look Once), Faster RCNN (Region-based Convolutional Neural Networks)) perform well but are sensitive to degraded images (haze, low light), reducing accuracy. To address blurred smoke features and attenuated flame brightness in degraded images, this paper proposes CoDeF-Net (Collaborative Detection Framework Network), a collaborative detection framework integrating Retinex-BCE (Retinex-based Bright Channel Enhancement) image enhancement with YOLOv11 (You Only Look Once version 11) to improve robustness. Experiments on 1757 real forest fire images show that Retinex-BCE achieves an FSIMC (Full-Reference Image Quality Assessment Metric based on Structural Similarity and Contrast) index of 0.9611 and an LOE (Loss of Edge) value of 254.78, preserving image structure. CoDeF-Net reaches AP@0.5 (Average Precision at Intersection over Union threshold 0.5) of 87.9% (3.8% higher than original YOLOv11), with low missed detection of small flames and enhanced stability in extreme scenarios, providing a feasible solution for forest fire monitoring under degraded images. Full article

Image enhancement is of pivotal importance in the detection of defects in wood veneers. However, acquired images frequently exhibit signs of blurring, uneven illumination, and insufficient contrast, which can lead to a reduction in the accuracy of defect recognition. In this study, an algorithm based on Grünwald–Letnikov (G–L) fractional-order differentiation is proposed for the enhancement of wood veneer defect images. Initially, the gain characteristics of differential amplitude-frequency responses on high- and low-frequency image components are analyzed, and the feasibility of the method is demonstrated by linking these characteristics with the frequency-domain distributions of live knot, dead knot, and crack defects. Secondly, an eight-direction mask operator is constructed based on the G–L definition, and a DC component preservation factor is introduced to eliminate the luminance drift caused by mask truncation. The application of the mask is performed independently on the R, G, and B channels, and a dynamic blending mechanism is designed to achieve a balance between texture enhancement and structural fidelity. Finally, a set of six evaluation metrics (AG, E, PSNR, RMSE, SSIM, and VIF) is employed to assess the quality of enhanced images. The proposed algorithm is then compared with five existing algorithms (SSR, MSR, MSRCR, CLAHE, and AGC) under both noise-free and additive white Gaussian noise conditions. The findings indicate that the G–L fractional-order differentiation algorithm facilitates a more balanced representation of image features, thereby enhancing contrast, brightness, and textural contours. This approach results in more authentic color reproduction and superior visual quality. Full article

Passive millimeter-wave (PMMW) imaging technology has become a highly promising technology that can protect privacy in human body security inspections. However, most existing methods rely on single-pixel and single-polarization processing mechanisms, which often lead to discrete false-alarm pixels or missed detections in practical applications. Although multi-polarization information can provide richer distinguishing features, the current methods typically depend on limited Stokes parameters or artificially designed polarization features, lacking a systematic framework to fully exploit the intrinsic potential of multi-polarization information. In this paper, we propose a novel multi-scale edge-preserving decomposition model, termed Gaussian and weighted average curvature filtering (GWACF), to hierarchically decompose a multi-polarization PMMW image into three structural layers: base structural (BS) layer, coarse structural (CS) layer, and fine structural (FS) layer. Furthermore, we also propose a fusion strategy in which a gradient-domain pulse-coupled neural network (PCNN) is employed to fuse the texture-rich CS and FS layers, while the energy attribute fusion method is applied to the BS layer where primary structure and background information play a dominant role. This method effectively leverages complementary polarimetric information without introducing artifacts or compromising edge sharpness. Experimental results demonstrate that the proposed method effectively enhances the brightness temperature (BT) contrast of concealed objects. Compared with existing mainstream methods, it exhibits notable advantages in both detection accuracy and robustness. Full article

Accurate oil–water interface measurement in small transparent test tubes is important for subsequent volume readout in laboratory analysis. However, manual observation and conventional vision-based methods are easily affected by illumination variation, wall stains, and bubbles, while deep learning detectors alone usually provide only coarse semantic perception. To address this issue, a coarse-to-fine framework is proposed for robust oil–water interface measurement. In the coarse stage, YOLOv8n is used to provide semantic constraints for subsequent processing. In the fine stage, a Fisher-discriminative chromatic-weighted brightness feature is constructed from RGB information, where the RGB weights are derived from the Fisher criterion to enhance oil–water chromatic separability rather than using fixed grayscale or empirical channel weights. This feature is then fused with a SobelY-based vertical-gradient feature to improve interface localization. A stain-aware row-aggregation strategy with effective-pixel compensation is further introduced to suppress artefact interference. The validated interface position is finally converted into a volume readout, with additional correction for bubble-induced bias. The framework was validated on sampled frames from a complete shale-oil core pressing process conducted under mixed-lighting conditions. Stage-wise evaluation and ablation results indicate that the proposed design improves readout stability under stains, bubbles, and illumination variation, achieving a mean absolute error of 0.0159 mL and keeping the maximum error below 0.03 mL in the current experimental setup. Full article

Visible and long-wave infrared (LWIR) image fusion is essential for robust imaging, yet balancing natural color and thermal detail remains challenging. This study proposes a novel framework integrating a selective pre-enhancement module, an advanced fusion network, and a Palette Denoising Diffusion Probabilistic Model (DDPM) for high-quality synthesis in no-reference environments. The fusion network employs an encoder–decoder architecture with Residual-in-Residual Dense Blocks (RRDBs) and a Convolutional Block Attention Module (CBAM) to extract discriminative multi-level features. Spatially Adaptive Normalization (SPADE) and a Visibility Deficiency Mask (VDF) are incorporated to adaptively compensate for information-poor regions while preserving modality-specific characteristics. The fused output is subsequently refined by a conditional DDPM to restore fine-grained textures and suppress noise. Finally, post-processing enhances global contrast and color naturalness while mitigating thermal artifacts. Experimental results demonstrate that the proposed method effectively reduces over-brightness and improves detail preservation in diverse nighttime scenarios. Full article

The solar corona has an extremely low density, and its brightness is only about one millionth of that of the photosphere. High-dynamic-range imaging of its faint structure is therefore essential for studying coronal heating, coronal mass ejections, and space weather. Quantitative coronagraph imaging requires flat-field measurement and calibration, which underpin intensity calibration, small-scale feature detection, and long-term cyclic analysis. This paper analyzes the coronagraph imaging chain (baffle–optical system–detector) and the origins of flat-field errors, including optical aberrations, stray light, and pixel-response non-uniformity, and summarizes the resulting calibration requirements of next-generation coronagraphs. On this basis, ground-based and space-based flat-fielding methods are systematically reviewed: the ground-based methods include integrating-sphere uniform light sources, opal glass/diffuser plates, clear-sky and thin-cloud backgrounds, and solar disk scanning, while the space-based methods include internal light sources and diffuser plates, attitude-roll and off-corona offset observations, and multi-phase statistical self-consistent flat-fielding. Their accuracy, resource cost, and applicability are compared. The review shows that no single method is simultaneously high-precision, easy to update, and engineer-friendly; a hierarchical, multi-method calibration framework is therefore recommended. Finally, a new method is proposed in which lithographically generated structured light fields, combined with Fourier optics and machine learning inversion, are used to estimate the pixel-response function. Preliminary experiments show that this method achieves a lower residual error than the integrating-sphere and opal glass methods, providing a high-precision reference for future wide-band, high-resolution coronagraph calibration. Full article