Search Results (1,514)

Early detection of small-scale fires is crucial for minimizing damage and enabling rapid emergency response. While recent deep learning-based fire detection systems have achieved high accuracy, they still face three key challenges: (1) limited deployability in resource-constrained edge environments due to high computational costs, (2) performance degradation caused by feature interference when jointly learning flame and smoke features in a single backbone, and (3) low sensitivity to small flames and thin smoke in the initial stages. To address these issues, we propose a lightweight dual-stream fire detection architecture based on YOLOv5n, which learns flame and smoke features separately to improve both accuracy and efficiency under strict edge constraints. The proposed method integrates two specialized attention modules: ESCFM++, which enhances spatial and channel discrimination for sharp boundaries and local flame structures (flame), and ESCFM-RS, which captures low-contrast, diffuse smoke patterns through depthwise convolutions and residual scaling (smoke). On the D-Fire dataset, the flame detector achieved 74.5% mAP@50 with only 1.89 M parameters, while the smoke detector achieved 89.2% mAP@50. When deployed on an NVIDIA Jetson Xavier NX(NVIDIA Corporation, Santa Clara, CA, USA)., the system achieved 59.7 FPS (single-stream) and 28.3 FPS (dual-tream) with GPU utilization below 90% and power consumption under 17 W. Under identical on-device conditions, it outperforms YOLOv9t and YOLOv12n by 36–62% in FPS and 0.7–2.0% in detection accuracy. We further validate deployment via outdoor day/night long-range live-stream tests on Jetson using our flame detector , showing reliable capture of small, distant flames that appear as tiny cues on the screen, particularly in challenging daytime scenes. These results demonstrate overall that modality-specific stream specialization and ESCFM attention reduce feature interference while improving detection accuracy and computational efficiency for real-time edge-device fire monitoring. Full article

The geographic distribution patterns of microorganisms and their underlying mechanisms are central topics in microbiology, crucial for understanding ecosystem functioning and predicting responses to global change. Cryoconite absorbs solar radiation to form cryoconite holes, and because it lies within these relatively deep holes, it faces limited interference from surrounding ecosystems, often being seen as a fairly enclosed environment. Moreover, it plays a dominant role in the biogeochemical cycling of key elements such as carbon and nitrogen, making it an ideal model for studying large-scale microbial biogeography. In this study, we analyzed bacterial communities in cryoconite across a transcontinental scale of glaciers to elucidate their biogeographical distribution and community assembly processes. The cryoconite bacterial communities were predominantly composed of Proteobacteria, Cyanobacteria, Bacteroidota, and Actinobacteriota, with significant differences in species composition across geographical locations. Bacterial diversity was jointly driven by geographical and anthropogenic factors: species richness exhibited a hump-shaped relationship with latitude and was significantly positively correlated with the Human Development Index (HDI). The significant positive correlation may stem from nutrient input and microbial dispersal driven by high-HDI regions’ industrial, agricultural, and human activities. Beta diversity demonstrated a distance-decay pattern along spatial gradients such as latitude and geographical distance. Analysis of community assembly mechanisms revealed that stochastic processes predominated across continents, with a notable scale dependence: as the spatial scale increased, the role of deterministic processes (heterogeneous selection) decreased, while stochastic processes (dispersal limitation) strengthened and became the dominant force. By integrating geographical, climatic, and anthropogenic factors into a unified framework, this study enhances the understanding of the spatial-scale-driven mechanisms shaping cryoconite bacterial biogeography and emphasizes the need to prioritize anthropogenic influences to predict the trajectory of cryosphere ecosystem evolution under global change. Full article

In citrus cultivation practice, regular monitoring of leaf leaf relative water content (RWC) can effectively guide water management, thereby improving fruit quality and yield. When applying hyperspectral technology to citrus leaf moisture monitoring, the precise quantification of RWC still needs to address issues such as data noise and algorithm adaptability. The noise interference and spectral aliasing in RWC sensitive bands lead to a decrease in the accuracy of moisture inversion in hyperspectral data, and the combined sensitive bands of chlorophyll (LCC) in citrus leaves can affect its estimation accuracy. In order to explore the optimal prediction model for RWC of citrus leaves and accurately control irrigation to improve citrus quality and yield, this study is based on 401–2400 nm spectral data and extracts noise robust features through continuous wavelet transform (CWT) multi-scale decomposition. A high-precision estimation model for citrus leaf RWC is established, and the potential of CWT in RWC quantitative inversion is systematically evaluated. This study is based on the multi-scale analysis characteristics of CWT to probe the time–frequency characteristic patterns associated with RWC and LCC in citrus leaf spectra. Pearson correlation analysis is used to evaluate the effectiveness of features at different decomposition scales, and the successive projections algorithm (SPA) is further used to eliminate band collinearity and extract the optimal sensitive band combination. Finally, based on the selected RWC and LCC-sensitive bands, a high-precision predictive model for citrus leaf RWC was established using partial least squares regression (PLSR). The results revealed that (1) CWT preprocessing markedly boosts the estimation accuracy of RWC and LCC relative to the original spectrum (max improvements: 6% and 3%), proving it enhances spectral sensitivity to these two indices in citrus leaves. (2) Combining CWT and SPA, the resulting predictive model showed higher inversion accuracy than the original spectra. (3) Integrating RWC Scale7 and LCC Scale5-2224/2308 features, the CWT-SPA fusion model showed optimal predictive performance (R² = 0.756, RMSE = 0.0214), confirming the value of multi-scale feature joint modeling. Overall, CWT-SPA coupled with LCC spectral traits can boost the spectral response signal of citrus leaf RWC, enhancing its prediction capability and stability. Full article

Considering the challenges associated with implementing emerging technologies and bacterial-derived antimicrobial metabolites at an industrial scale in the meat industry, this comprehensive review investigates the interactions between lactic acid bacteria-producing antimicrobial metabolites and emerging food preservation technologies applied to meat systems. By integrating evidence from microbiology, food engineering, and molecular physiology, the review characterizes how metabolites-derived compounds exert inhibitory activity through pH modulation, membrane permeabilization, disruption of proton motive force, and interference with cell wall biosynthesis. These biochemical actions are evaluated in parallel with the mechanistic effects of high-pressure processing, pulsed electric fields, cold plasma, irradiation, pulsed light, ultrasound, ohmic heating and nanotechnology. Across the literature, consistent patterns of synergy emerge: many emerging technologies induce structural and metabolic vulnerabilities in microbial cells, thereby amplifying the efficacy of antimicrobial metabolites while enabling reductions in process intensity. The review consolidates these findings to elucidate multi-hurdle strategies capable of improving microbial safety, extending shelf life, and preserving the physicochemical integrity of meat products. Remaining challenges include optimizing combinational parameters, ensuring metabolite stability within complex matrices, and aligning integrated preservation strategies with regulatory and industrial constraints. Full article

The floodplain landscape of the lower Yellow River is jointly shaped by natural water-sediment processes and human activities. With intensified regulation by large reservoirs and increasing human development intensity, the landscape pattern of the floodplain has undergone significant changes. Clarifying the relative contributions of natural and anthropogenic factors, as well as their interactive mechanisms, is crucial for ecological management of the floodplain. Based on 40-year long-term land-use data and hydrological and meteorological observations, this study integrates landscape metrics, the human interference index (HI), grey relational analysis, and partial least squares regression to quantify the spatiotemporal dynamics of landscape pattern in the floodplain of the lower Yellow River and to elucidate the driving mechanisms underlying landscape-pattern evolution. The results indicate that (1) during the study period, the areas of cultivated land and built-up land in the floodplain continuously increased, whereas natural wetlands and grassland decreased accordingly. Taking 2000 as a breakpoint, the rate and direction of landscape change exhibited stage-dependent differences. (2) Landscape pattern metrics changed nonlinearly: the number of patches decreased first and then increased; the patch cohesion index increased first and then declined; and Shannon’s diversity index showed an overall downward trend. These changes suggest a process of landscape consolidation induced by agricultural cultivation, followed by re-fragmentation driven by the expansion of built-up land. (3) Driving-mechanism analysis shows that the HI is the primary driver of the current changes in floodplain landscape pattern. After the operation of the Xiaolangdi Dam, water-sediment conditions tended to stabilize and flood risk in the floodplain decreased, thereby creating favourable conditions for human activities. This study highlights the stage-dependent influences of natural and anthropogenic factors on floodplain landscape evolution under dam regulation and suggests that management strategies should be adapted to the current re-fragmentation phase, prioritizing the strict control of agricultural expansion and the restoration of ecological corridors to mitigate anthropogenic interference under stable dam regulation. Full article

To address the challenging requirements of citrus detection in complex orchard environments, this paper proposes ACDNet (Adaptive Citrus Detection Network), a novel deep learning framework specifically designed for automated citrus harvesting. The proposed method introduces three key innovations: (1) Citrus-Adaptive Feature Extraction (CAFE) module that combines fruit-aware partial convolution with illumination-adaptive attention mechanisms to enhance feature representation with improved efficiency; (2) Dynamic Multi-Scale Sampling (DMS) operator that adaptively focuses sampling points on fruit regions while suppressing background interference through content-aware offset generation; and (3) Fruit-Shape Aware IoU (FSA-IoU) loss function that incorporates citrus morphological priors and occlusion patterns to improve localization accuracy. Extensive experiments on our newly constructed CitrusSet dataset, which comprises 2887 images capturing diverse lighting conditions, occlusion levels, and fruit overlapping scenarios, demonstrate that ACDNet achieves superior performance with mAP@0.5 of 97.5%, precision of 92.1%, and recall of 92.8%, while maintaining real-time inference at 55.6 FPS. Compared to the baseline YOLOv8n model, ACDNet achieves improvements of 1.7%, 3.4%, and 3.6% in mAP@0.5, precision, and recall, respectively, while reducing model parameters by 11% (to 2.67 M) and computational cost by 20% (to 6.5 G FLOPs), making it highly suitable for deployment in resource-constrained robotic harvesting systems. However, the current study is primarily validated on citrus fruits, and future work will focus on extending ACDNet to other spherical fruits and exploring its generalization under extreme weather conditions. Full article

Unmanned Aerial Vehicle (UAV) visual tracking is widely used but continues to face challenges such as unpredictable target motion, error accumulation, and the sparse appearance of small targets. To address these issues, we propose a Mamba-based Spatio-Temporal Fusion Tracker. To address tracking drift from large displacements and abrupt pose changes, we first introduce a Bidirectional Spatio-Temporal Mamba module. It employs bidirectional spatial scanning to capture discriminative local features and temporal scanning to model dynamic motion patterns. Second, to suppress error accumulation in complex scenes, we develop a Dynamic Template Fusion module with Adaptive Attention. This module integrates a threefold safety verification mechanism—based on response peak, temporal consistency, and motion stability—with a scale-aware strategy to enable robust template updates. Moreover, we design a Small-Target-Aware Context Prediction Head that utilizes a Gaussian-weighted prior to guide feature fusion and refines the loss function, significantly improving localization accuracy under sparse target features and strong background interference. On three major UAV tracking benchmarks (UAV123, UAV123@10fps, and UAV20L), our MSTFT establishes new state-of-the-art with success AUCs of 79.4%, 76.5%, and 75.8% respectively. More importantly, it maintains a tracking speed of 45 FPS, demonstrating a superior balance between precision and efficiency. Full article

The data collection and analytical capabilities of the Instrumentation and Control (I&C) system in nuclear power plants (NPPs) continue to advance, thereby enhancing operational state awareness and enabling more precise control. However, the data acquisition, transmission, and storage devices in nuclear power plant (NPP) I&C systems typically operate in harsh environments. This exposure can lead to device failures and susceptibility to external interference, potentially resulting in data anomalies such as missing samples, signal skipping, and measurement drift. This paper presents a Gated Recurrent Unit and Multilayer Perceptron (GRU-MLP)-based method for anomaly detection and correction in NPP I&C system data. The goal is to improve operational data quality, thereby supplying more reliable input for system analysis and automatic controllers. Firstly, the short-term prediction algorithm of operation data based on the GRU model is studied to provide a reference for operation data anomaly detection. Secondly, the MLP model is connected to the GRU model to recognize the difference between the collected value and the prediction value so as to distinguish and correct the anomalies. Finally, a series of experiments were conducted using operational data from a pressurized water reactor (PWR) to evaluate the proposed method. The experiments were designed as follows: (1) These experiments assessed the model’s prediction performance across varying time horizons. Prediction steps of 1, 3, 5, 10, and 20 were configured to verify the accuracy and robustness of the data prediction capability over short and long terms. (2) The model’s effectiveness in identifying anomalies was validated using three typical patterns: random jump, fixed-value drift, and growth drift. The growth drift category was further subdivided into linear, polynomial, and logarithmic growth to comprehensively test detection performance. (3) A comparative analysis was performed to demonstrate the superiority of the proposed GRU-MLP algorithm. It was compared against the interactive window center value method and the ARIMA algorithm. The results confirm the advantages of the proposed method for anomaly detection, and the underlying reasons are analyzed. (4) Additional experiments were carried out to discuss and verify the mobility (or transferability) of the prediction algorithm, ensuring its applicability under different operational conditions. Full article

Morningglories (Ipomoea lacunosa, I. hederacea, and I. purpurea) are persistent, problematic weeds in summer row crops throughout warm-temperate regions. Their vining growth habit and enduring seedbanks lead to recurring infestations and harvest interferences. This review synthesizes current knowledge on the seed ecology of these species to clarify how dormancy, germination, and emergence processes contribute to their persistence. Published anatomical and ecological studies were examined to summarize dormancy mechanisms, environmental signals regulating dormancy release, germination requirements, and seasonal emergence patterns. Morningglories exhibit a dormancy system dominated by physical dormancy, occasionally combined with a transient physiological component. Dormancy release is promoted by warm and fluctuating temperatures, hydration–dehydration cycles, and long-term seed-coat weathering. Once permeable, seeds germinate across broad temperature ranges, vary in sensitivity to water potential, and show limited dependence on light. Field studies indicate extended emergence windows from late spring through midsummer, especially in no-till systems where surface seeds experience strong thermal and moisture fluctuations. Despite substantial progress, significant gaps remain concerning maternal environmental effects, population-level variation, seedbank persistence under modern management, and the absence of mechanistic emergence models. An improved understanding of these processes will support the development of more predictive and ecologically informed management strategies. Full article

Traffic flow prediction is a fundamental component of intelligent transportation systems and plays a critical role in traffic management and autonomous driving. However, accurately modeling highway traffic remains challenging due to dynamic congestion propagation, lane-level heterogeneity, and non-recurrent traffic events. To address these challenges, this paper proposes an improved attention-mechanism spatio-temporal graph convolutional network, termed AMSGCN, for highway traffic flow prediction. AMSGCN introduces an adaptive adjacency matrix learning mechanism to overcome the limitations of static graphs and capture time-varying spatial correlations and congestion propagation paths. A hierarchical multi-scale spatial attention mechanism is further designed to jointly model local congestion diffusion and long-range bottleneck effects, enabling an adaptive spatial receptive field under congested conditions. To enhance temporal modeling, a gating-based fusion strategy dynamically balances periodic patterns and recent observations, allowing effective prediction under both regular and abnormal traffic scenarios. In addition, direction-aware encoding is incorporated to suppress interference from opposite-direction lanes, which is essential for directional highway traffic systems. Extensive experiments on multiple benchmark datasets, including PeMS and PEMSF, demonstrate the effectiveness and robustness of AMSGCN. In particular, on the I-24 MOTION dataset, AMSGCN achieves an RMSE reduction of 11.0% compared to ASTGCN and 17.4% relative to the strongest STGCN baseline. Ablation studies further confirm that dynamic and multi-scale spatial attention provides the primary performance gains, while temporal gating and direction-aware modeling offer complementary improvements. These results indicate that AMSGCN is a robust and effective solution for highway traffic flow prediction. Full article

The motivation for this work is to study the cause and present mitigation for some challenges faced in preparing porous silicon. This enables benefiting from the appealing benefits of porous silicon that offers a wide range, simple technique for varying the refractive index. Such challenges include the refractive index values, sensitivity to oxidation, some fabrication parameters, and other factors. Additionally, highly doped p-type silicon is preferred to form porous silicon, but it causes high losses, which necessitates its detachment. We investigate some possible causes of refractive index change, especially after detaching the fabricated layers from the silicon substrate. Thereby, we could recommend simple but essential precautions during fabrication to avoid such a change. For example, the native oxide formed in the pores has a role in changing the porosity upon following some fabrication sequence. Oppositely, intrinsic stress doesn’t have a significant role. On another aspect, the effect of differing etching/break times on the filter’s responses has been studied, along with other subtle details that may affect the lateral and depth homogeneity, and thereby the process success. Solving such homogeneity issues allowed reaching thick layers not suffering from the gradient index. It is worth highlighting that several approaches have been reported; unlike these, our method doesn’t require sophisticated equipment that might not be available in every lab. To well characterize the thin films, it has been found essential that freestanding monolayers are used for this purpose. From which, the wavelength-dependent refractive index and absorption coefficient have been determined in the near infrared region (1000–2500 nm) for different fabricated conditions. Excellent fitting with the measured interference pattern has been achieved, indicating the accurate parameter extraction, even without any ellipsometry measurements. This also demonstrates the refractive index homogeneity of the fabricated layer, even with a large thickness of over 16 µm. Subsequently, multilayer structures have been fabricated and tested, showing the successful nano-manufacturing methodology. Full article

As power systems expand and grow smarter, the safe and steady operation of substation equipment has become a prerequisite for grid reliability. In cluttered substation scenes, however, existing deep learning detectors still struggle with small targets, multi-scale feature fusion, and precise localization. To overcome these limitations, we introduce PBZGNet, a defect-detection network that couples a gradual parallel-branch backbone, a zoom-fusion neck, and a global channel-recalibration module. First, BiCoreNet is embedded in the feature extractor: dual-core parallel paths, reversible residual links, and channel recalibration cooperate to mine fault-sensitive cues. Second, cross-scale ZFusion and Concat-CBFuse are dynamically merged so that no scale loses information; a hierarchical composite feature pyramid is then formed, strengthening the representation of both complex objects and tiny flaws. Third, an attention-guided decoupled detection head (ADHead) refines responses to obscured and minute defect patterns. Finally, within the Generalized Focal Loss framework, a quality rating scheme suppresses background interference while distribution regression sharpens the localization of small targets. Across all scales, PBZGNet clearly outperforms YOLOv11. Its lightweight variant, PBZGNet-n, attains 83.9% mAP@50 with only 2.91 M parameters and 7.7 GFLOPs—9.3% above YOLOv11-n. The full PBZGNet surpasses the current best substation model, YOLO-SD, by 7.3% mAP@50, setting a new state of the art (SOTA). Full article

Timely tool wear detection has been an important target for the metal cutting industry for decades because of its significance for part quality and production cost control. With the shift toward intelligent and sustainable manufacturing, reliable tool-condition monitoring has become even more critical. One of the main challenges in sound-based tool wear monitoring is the presence of noise interference, instability and the highly volatile nature of machining acoustics, which complicates the extraction of meaningful features. In this study, a Convolutional Neural Network (CNN) model is proposed to classify tool wear conditions in milling operations using acoustic signals. Sound recordings were collected from tools at different wear stages under two cutting speeds, and Mel-Frequency Cepstral Coefficients (MFCCs) were extracted to obtain a compact representation of the short-term power spectrum. These MFCC matrices enabled the CNN to learn discriminative spectral patterns associated with wear. To evaluate model stability and reduce the effects of algorithmic randomness, training was repeated three times for each cutting speed. For the 520 rpm dataset, the model achieved an average validation accuracy of 96.85 ± 2.07%, while for the 635 rpm dataset it achieved 93.69 ± 2.07%. The results demonstrate the feasibility of using acoustic signals, despite inherent noise challenges, as a complementary approach for identifying suitable tool replacement intervals in milling. Full article

In the breeding scene, limited by the small number of samples and environmental interference such as illumination occlusion, sika deer behavior recognition still faces challenges such as insufficient feature representation and weak cross-scale modeling ability. To this end, this study builds a lightweight improved model SDB-YOLO based on YOLOv11n. Firstly, the FPSC module is proposed to enhance the correlation between multi-scale features through the shared convolution mechanism, so as to significantly improve the quality of feature fusion under the condition of small samples. Secondly, the Ghost feature generation and dynamic convolution strategy are introduced into the C3k2 module to construct the C3_GDConv structure, so as to strengthen the fine-grained behavior pattern modeling ability and reduce redundant calculations. In addition, the CBAM attention mechanism is added to the neck of the network to further improve the ability of key information extraction and enhance the discrimination of feature expression. Finally, the EfficientHead was used to replace the original detection head to obtain a more robust training process and higher detection accuracy in small-sample scenarios. Experimental results show that SDB-YOLO achieves 90.2% detection accuracy with only 4.3 GFLOPs of calculation, which achieves significant performance improvement compared with YOLOv11n, and verifies the effectiveness and lightweight advantages of the proposed method in small-sample special animal behavior recognition tasks. Full article

Accurate forecasting of metro energy consumption is essential for intelligent power management and sustainable urban transportation systems. However, existing studies often overlook the intrinsic properties of metro energy time series, such as strong periodicity, inter-line heterogeneity, and pronounced non-stationarity. To address this gap, this paper proposes an enhanced Informer-based framework, PatchConvFormer (PCformer). The model integrates three key innovations: (1) a channel-independent modeling mechanism that reduces interference across metro lines; (2) a patch-based temporal segmentation strategy that captures fine-grained intra-cycle energy fluctuations; and (3) a multi-scale convolution-augmented attention module that jointly models short-term variations and long-term temporal dependencies. Using real operation data from 16 metro lines in a major city in China, PCformer achieves significant improvements in forecasting accuracy (MSE = 0.043, MAE = 0.145). Compared with the strongest baseline model in each experiment (i.e., the second-best model), the MSE and MAE are reduced by approximately 41.9% and 19.8%, respectively. In addition, the model maintains strong stability and generalization across different prediction horizons and cross-line transfer experiments. The results demonstrate that PCformer effectively enhances Informer’s capability in modeling complex temporal patterns and provides a reliable technical framework for metro energy forecasting and intelligent power scheduling. Full article