Search Results (6,149)

Multi-object tracking aims to track multiple objects across consecutive frames in a video, assigning a unique classifier to each object. However, issues such as occlusions, directional changes, or shape alterations can cause appearance variations, leading to detection and matching problems that in turn result in frequent ID switches. To solve these issues, this paper proposes a two-stage multi-object tracking framework based on a spatial and temporal fusion algorithm. First, the video frames are processed by a detector to identify objects and form rectangular detection areas. Meanwhile, an estimator predicts the target rectangular areas in the next frame. Then, we extract the optical flow of the target pixels within the detection and prediction areas, and then a temporal information model is established by calculating the average of the target pixels’ optical flow. Afterward, we present a spatial information model using the R-IoU (Reverse of Intersection over Union) between the detection and prediction areas. This spatial and temporal information is combined with weighted matrix fusion, which achieves the feature matching and association task. Finally, we implement a two-stage association multi-object tracking model using the mentioned fusion algorithm. Experiments on the MOTChallenge dataset using the official detector show that our two-stage multi-object tracking method based on the spatial and temporal fusion algorithm is robust in handling occlusions and ID switch issues. As of the submission of this paper, the proposed method has achieved the top ranking in the MOT17 benchmark when evaluated with the official detector. Full article

With urbanization, ozone (O₃) pollution and the urban heat island (UHI) effect have become increasingly prominent. UHI can affect O₃ production and its dilution and dispersion, but the underlying mechanisms remain unclear. This study investigates the spatial and temporal distribution of O₃ pollution and the UHI effect, as well as the influence of UHI on O₃ pollution in the Sichuan Basin. Atmospheric pollution data for O₃ and NO₂ from 2020 were obtained from local environmental monitoring stations, while temperature and single-layer wind field data were sourced from ERA5-Land, a high-resolution atmospheric reanalysis dataset provided by the European Centre for Medium-Range Weather Forecasts (ECMWF). The results indicate the following: (1) O₃ concentrations in the Sichuan Basin exhibit distinct seasonal variations, with the highest levels in spring, followed by summer and autumn, and the lowest in winter. In terms of spatial variation, the overall distribution is highest in western Sichuan, second highest along the Sichuan River, and lowest in central Sichuan. (2) There are significant regional differences in UHII across Sichuan, with medium heat islands (78.63%) dominating western Sichuan, weak heat islands (82.74%) along the Sichuan River, and no heat island (34.79%) or weak heat islands (63.56%) in central Sichuan. Spatially, UHII is mainly distributed in a circular pattern. (3) Typical cities in the Sichuan Basin (Chengdu, Chongqing, Nanchong) show a positive correlation between UHII and O₃ concentration (0.071–0.499), though with an observed temporal lag. This study demonstrates that UHI can influence O₃ concentrations in two ways: first, by altering local heat balance, thereby promoting O₃ production, and second, by generating local winds that contribute to the diffusion or accumulation of O₃, forming distinct O₃ concentration zones. Full article

In this study, a diagnostic model for lower atmospheric ducts was developed using the polar weather research and forecasting model. A five-year simulation was then conducted across the entire Arctic region to investigate the temporal and spatial characteristics of lower atmospheric ducts. The model demonstrated excellent performance in simulating modified atmospheric refractivity, with root mean square errors ranging from 0 M to 5 M. The five-year simulation results revealed that duct occurrence rates across the Arctic region were all below 1% and exhibited a negative relationship with latitude. Regarding the difference between surface ducts and elevated ducts, a higher frequency of surface ducts was detected in the Arctic region. The height and thickness of surface ducts were generally lower than those of elevated ducts, but the strength of surface ducts was slightly greater. Regionally, surface ducts mainly occurred in the land areas surrounding the Arctic Ocean, while more elevated ducts were found in the North Atlantic Sea area. Additionally, a negative correlation was observed between the polar vortex indices and the characteristics of ducts, particularly for surface ducts. The ducts in Greenland were notably influenced by polar vortex activity, whereas the ducts in other regions, such as the Norwegian Sea and Kara Sea, were less affected. Full article

Crop phenotype detection is a precise way to understand and predict the growth of horticultural seedlings in the smart agriculture era to increase the cost-effectiveness and energy efficiency of agricultural production. Crop phenotype detection requires the consideration of plant stature and agricultural devices, like robots and autonomous vehicles, in smart greenhouse ecosystems. However, collecting the imaging dataset is a challenge facing the deep learning detection of plant phenotype given the dynamic changes among leaves and the temporospatial limits of camara sampling. To address this issue, digital cousin is an improvement on digital twins that can be used to create virtual entities of plants through the creation of dynamic 3D structures and plant attributes using RGB image datasets in a simulation environment, using the principles of the variations and interactions of plants in the physical world. Thus, this work presents a two-phase method to obtain the phenotype of horticultural seedling growth. In the first phase, 3D Gaussian splatting is selected to reconstruct and store the 3D model of the plant with 7000 and 30,000 training rounds, enabling the capture of RGB images and the detection of the phenotypes of the seedlings, overcoming temporal and spatial limitations. In the second phase, an improved YOLOv8 model is created to segment and measure the seedlings, and it is modified by adding the LADH, SPPELAN, and Focaler-ECIoU modules. Compared with the original YOLOv8, the precision of our model is 91%, and the loss metric is lower by approximately 0.24. Moreover, a case study of watermelon seedings is examined, and the results of the 3D reconstruction of the seedlings show that our model outperforms classical segmentation algorithms on the main metrics, achieving a 91.0% mAP50 (B) and a 91.3% mAP50 (M). Full article

Exploring the characteristics and driving factors of carbon storage change in different terrain gradient variations can provide important insights for formulating the agricultural ecological protection policy for regional development. Previous studies have used the fixed value of carbon density to evaluate the change characteristics of carbon storage but ignored the spatio-temporal heterogeneity of carbon storage at the block scale and the impact of policy factors. Thus, this paper takes Sanjiang Plain, Heilongjiang Province, China, as a study area, and the spatio-temporal variation of carbon storage at different topographic gradients was revealed using hot and cold spot analysis and zonal statistics. Through the geographic detector and estimation of the soil carbon density model, the driving factors and intensity of carbon storage spatial distribution are revealed from 1990 to 2020. We conducted analyses on aboveground biomass, underground biomass, and soil carbon storage across three elevation levels (0–200 m, 200–500 m, 500–999 m) to reveal the quantitative distribution features of carbon storage. The study analysis finds that carbon storage indicates a sawtooth evolution during the study period. Carbon storage was dominant at elevation I (range is 0–200 m), slope I (range is 0–2°), and relief amplitude I (range is 0–30 m). Additionally, the carbon storage losses were severe at elevation II (range is 200–500 m), slope II (2–6°), and relief amplitude II (30–70 m). In contrast, the carbon storage losses at elevation III (500–999 m), slope III (6–15°), and relief amplitude III (70–186 m) were insignificant. The spatial pattern of carbon storage varies significantly under different topographic gradients from 1990 to 2020. The most critical driving factors influencing the spatial distribution pattern of carbon storage were land use and annual average temperature. Distance to urban centers and soil texture also moderately influence the distribution of carbon storage. As the topographic gradient increases, the dominant factors of carbon storage gradually change from annual mean temperature and the extent of land use to policy factors and other socio-economic factors. Therefore, this study emphasizes the importance of implementing policies that convert farmland to forests and wetlands and promote the green transformation of agriculture. Full article

This study examines the spatio-temporal variations of ambient air pollution from fine particulates below 2.5 µm (PM_2.5) and particulate matter below 10 µm (PM₁₀) in three of the most polluted provinces in Iran, namely Tehran, Isfahan, and Khuzestan, over a 6-year period (2016–2021). The results reveal distinct patterns of PM₁₀ and PM_2.5 concentrations since in Tehran, the highest PM₁₀ and PM_2.5 levels occur in winter, while PM_2.5 is lowest from March to May. Khuzestan experiences the highest pollution levels in summer due to dust storms, while Isfahan exhibits pollution levels and annual patterns similar to Tehran. Strong correlations are observed between PM₁₀ and PM_2.5 concentrations at stations in Tehran and Khuzestan Provinces, suggesting common sources and variation in both coarse and fine PM, with average PM_2.5/PM₁₀ ratios of 0.39–0.42, suggesting the dominance of dust. Furthermore, the analysis identifies the role of atmospheric stability, wind speed, and dust storms in controlling the PM levels in the three provinces. Lifetime cancer risks have been identified as unacceptably high, exceeding the threshold limit of 10⁻⁴, while Hazard Quotient (HQ) values above 1 indicate a high non-carcinogenic potential risk, particularly at stations in Khuzestan Province. The Excess Lifetime Cancer Risk (ELCR) values for PM_2.5 exposure in the most populated Tehran Province range from 139.4 × 10⁻⁶ to 263.2 × 10⁻⁶, underscoring significant cancer risks across various monitoring sites. This study emphasizes the urgent need for targeted pollution control measures in each province to effectively mitigate the adverse health effects associated with high PM concentrations. Full article

Epilepsy is a group of neurological disorders characterized by epileptic seizures, and it affects tens of millions of people worldwide. Currently, the most effective diagnostic method employs the monitoring of brain activity through electroencephalogram (EEG). However, it is critical to predict epileptic seizures in patients prior to their onset, allowing for the administration of preventive medications before the seizure occurs. As a pivotal application of artificial intelligence in medical treatment, learning the features of EEGs for epilepsy prediction and detection remains a challenging problem, primarily due to the presence of intra-class and inter-class variations in EEG signals. In this study, we propose the spatio-temporal EEGNet, which integrates contractive slab and spike convolutional deep belief network (CssCDBN) with a self-attention architecture, augmented by dual-task learning to address this issue. Initially, our model was designed to extract high-order and deep representations from EEG spectrum images, enabling the simultaneous capture of spatial and temporal information. Furthermore, EEG-based verification aids in reducing intra-class variation by considering the time correlation of the EEG during the fine-tuning stage, resulting in easier inference and training. The results demonstrate the notable efficacy of our proposed method. Our method achieved a sensitivity of 98.5%, a false-positive rate (FPR) of 0.041, a prediction time of 50.92 min during the epilepsy prediction task, and an accuracy of 94.1% during the epilepsy detection task, demonstrating significant improvements over current state-of-the-art methods. Full article

Exploring and predicting the spatiotemporal evolution characteristics and driving forces of carbon storage in typical mountain forest ecosystems under land-use changes is crucial for curbing the effects of climate change and fostering sustainable, eco-friendly growth. The existing literature provides important references for our related studies but further expansion and improvements are needed in some aspects. This study first proposed an integrated framework comprising gray multi-objective optimization (GMOP), Integrated Valuation of Ecosystem Services and Tradeoffs (InVEST), the Patch-level Land Use Simulation Model (PLUS), and optimal parameter-based geographical detector (OPGD) models to further expand and improve on existing research. Then, the integrated model was used to analyze the spatial–temporal variation in land-use pattern and carbon storage at the county scale in China’s Daiyun Mountain’s Rim under four scenarios in 2032, and analyze the driving force of spatial differentiation of carbon storage. The results indicated that (1) land-use change primarily involves the mutual transfer among forest, cultivated, and construction land, with approximately 7.2% of the land-use type area undergoing a transition; (2) in 2032, the natural development scenario projects a significant reduction in forest land and an expansion of cultivated, shrub, and construction lands. Conversely, the economic priority, ecological priority, and economic–ecological coordinated scenarios all anticipate a decline in cultivated land area; (3) in 2032, the natural development scenario will see a 2.8 Tg drop in carbon stock compared to 2022. In contrast, the economic priority, ecological priority, and economic–ecological coordinated scenarios are expected to increase carbon storage by 0.29 Tg, 2.62 Tg, and 1.65 Tg, respectively; (4) the spatial differentiation of carbon storage is jointly influenced by various factors, with the annual mean temperature, night light index, elevation, slope, and population density being the key influencing factors. In addition, the influence of natural factors on carbon storage is diminishing, whereas the impact of socioeconomic factors is on the rise. This study deepened, to a certain extent, the research on spatiotemporal dynamics simulation of carbon storage and its driving mechanisms under land-use changes in mountainous forest ecosystems. The results can serve to provide scientific support for carbon balance management and climate adaptation strategies at the county scale while also offering case studies that can inform similar regions around the world. However, several limitations remain, as follows: the singularity of carbon density data, and the research scope being confined to small-scale mountainous forest ecosystems. Future studies could consider collecting continuous annual soil carbon density data and employing land-use simulation models (such as PLUS or CLUMondo) appropriate to the study area’s dimensions. Full article

To effectively monitor the nonlinear wear variation of tools during the processing of titanium alloys, this study proposes a hybrid deep neural network fault diagnosis model that integrates the triangulation topology aggregation optimizer (TTAO), convolutional neural network (CNN), bidirectional long short-term memory network (BiLSTM), and attention mechanism (AM). Firstly, vibration signals from the machine tool spindle are acquired and subjected to the wavelet packet transform (WPT) to extract multi-frequency band energy features as model inputs. Then, the CNN and BiLSTM modules capture the features and temporal relationships of the input signals. Finally, introduction of the AM, combined with the TTAO algorithm, automatically extracts deep features, overcoming issues such as local optima and slow convergence in traditional neural networks, thereby enhancing the accuracy and efficiency of tool wear state recognition. The experimental results demonstrate that the proposed model achieves an average accuracy rate of 98.649% in predicting tool wear states, outperforming traditional backpropagation (BP) networks and standard CNN models. Full article

Globally, there has been a lot of focus on climate variability, especially variability in annual precipitation and temperatures. Depending on the area, different climate variables have different degrees of variation. Therefore, analyzing the temporal and spatial changes or dynamics of meteorological or climatic variables in light of climate change is crucial to identifying the changes induced by climate and providing workable adaptation solutions. This study examined how climate variability affects tea production in Darjeeling, West Bengal, India. It also looked at trends in temperature and rainfall between 1991 and 2023. In order to identify significant trends in these climatic factors and their relationship to tea productivity, the study used a variety of statistical tests, including the Sen’s Slope Estimator test, the Mann–Kendall’s test, and regression tests. The study revealed a positive growth trend in rainfall (Sen’s slope = 0.25, p = 0.001, R² = 0.032), maximum temperature (Sen’s slope = 1.02, p = 0.026, R² = 0.095), and minimum temperature (Sen’s slope = 4.38, p = 0.006, R² = 0.556). Even with the rise in rainfall, there has been a decline in tea productivity, as seen by the sharp decline in both the tea cultivated area and the production of tea. The results obtained from the regression analysis showed an inverse relationship between temperature anomalies and tea yield (R = −0.45, p = 0.02, R² = 0.49), indicating that the growing temperatures were not favorable for the production of tea. Rainfall anomalies, on the other hand, positively correlated with tea yield (R = 0.56, p = 0.01, R² = 0.68), demonstrating that fluctuations in rainfall have the potential to affect production but not enough to offset the detrimental effects of rising temperatures. These results underline how susceptible the tea sector in Darjeeling is to climate change adversities and the necessity of adopting adaptive methods to lessen these negative consequences. The results carry significance not only for regional stakeholders but also for the global tea industry, which encounters comparable obstacles in other areas. Full article

Methane (CH₄), the second-largest greenhouse gas contributing to global warming, has a high warming potential despite its short atmospheric lifespan. Inner Mongolia, due to its high carbon and energy consumption industries, faces significant methane emission challenges. This study uses TROPOMI satellite data (February 2019 to December 2022) to analyze the long-term trends and spatial distribution of methane in Inner Mongolia. The results indicate significant spatial heterogeneity in the methane concentration distribution in Inner Mongolia, China. Higher methane concentrations are observed in the southeastern regions, whereas the central regions exhibit relatively lower concentrations. Temporally, the methane concentrations show an increasing trend with seasonal peaks from late August to early September. Using multiple stepwise regression and geographically weighted regression (GWR) methods, the study identifies the key factors influencing methane concentrations. Increased precipitation and soil temperature, along with intensified human activity, contribute to higher methane levels, while rising surface temperatures and increased vegetation suppress methane concentrations. The GWR model provides a better fit compared to the traditional methods, especially in regions with higher methane levels. This research offers insights for developing strategies to mitigate methane emissions and supports China’s emission control targets. Full article

Heavy metal contamination in agricultural soils has garnered increasing attention, yet research on the spatiotemporal trends of heavy metal pollution in tropical regions with multiple annual crop harvests remains limited. This study examines data from 41 studies published between 2000 and 2024, including 206 records from 4122 sampling points on Hainan Island in China, to investigate the spatial distribution and temporal trends of heavy metal pollution. The results reveal that the average concentrations of Cd, Pb, As, Cr, and Hg in surface soil samples from agricultural lands on Hainan Island are 0.12, 28.28, 4.36, 63.98, and 0.075 mg/kg, respectively, all below the risk screening thresholds set by the Soil Pollution Risk Control Standard for Agricultural Land (GB 15618-2018). Spatially, heavy metal concentrations exhibit considerable regional variation. Cd levels are lower in the central region but higher in the northern and southern parts of the island. Both the cumulative pollution index and potential ecological risk index are elevated at the northern and southern ends, indicating more severe pollution in these areas. Pb and As show similar spatial patterns, with higher concentrations in the west and lower concentrations in the east. Conversely, Cr has higher concentrations in the northeast and lower concentrations in the southwest. Hg levels are elevated at the northern and southern ends of the island, though the overall pollution and ecological risk in these areas remain relatively low. Temporally, the concentration of heavy metals in agricultural soils has increased overall over the past two decades, with peak values occurring between 2017 and 2023. From 2002 to 2013, the variation was modest, while the largest fluctuations occurred between 2014 and 2016. Among the metals, Cr exhibited the most significant increase, indicating the most severe pollution, followed by Cd and Hg. As and Pb showed relatively lower levels of contamination. Regarding exceedance rates, the exceedances were evaluated against the thresholds established in GB15618-2018 and GB15618-1995. Cd’s exceedance rate increased from approximately 1% between 2002 and 2014 to between 7.78% and 20.93% in the following years, peaking in 2017. The exceedance rate for As rose slightly from 0% to 0.83%, with sporadic exceedances starting in 2015. Although these were relatively minor, a severe pollution point for As was observed in 2019. Exceedance rates for Pb and Cr increased significantly, from 0.75% and 7.50% in 2019 to 1.94% and 9.44% in 2023, reflecting increases of 4.8 to 10 times. These findings underscore the need for strengthened monitoring and management of heavy metal pollution in agricultural soils on Hainan Island to safeguard land quality and ensure the sustainability of local agricultural practices. Full article

Despite Guangxi’s unique ecological diversity and its important role in land-based ecological security and conservation, research on the assessment and prediction of its habitat quality under the influences of rapid urbanization and environmental pressures remains limited. This study systematically analyzes the spatial and temporal dynamics of land use and habitat quality in Guangxi from 2000 to 2020 using the PLUS-InVEST model and simulates future scenarios for 2030. These scenarios include the Natural Development (ND) scenario, Urban Development (UD) scenario, and Cropland and Ecological Protection (CE) scenario. The results indicate the following: (1) Over the past two decades, rapid urban and construction land expansions in Guangxi intensified their negative impact on habitat degradation. Additionally, the disproportionate change between rural settlement land and rural population warrants attention. (2) Although ecological restoration measures have played a positive role in mitigating habitat degradation, their effects have been insufficient to counterbalance the negative impacts of construction land expansion, highlighting the need for balanced land use planning and urbanization policies. (3) The expansion of rural residential areas had a greater impact on regional habitat quality degradation than urban and infrastructure expansion. Moderate urbanization may contribute to habitat quality improvement. (4) The CE scenario shows the most significant improvement in habitat quality (an increase of 0.13%), followed by the UD scenario, which alleviates habitat degradation by reducing pressure on rural land. In contrast, the ND scenario predicts further declines in habitat quality. Furthermore, land use planning, restoration measures, and sustainable development policies are key factors influencing habitat quality changes. These findings emphasize the importance of integrating land use strategies with ecological restoration measures to balance economic growth and biodiversity conservation, especially in rapidly urbanizing regions. Full article

Studying the temporal and spatial variation characteristics and driving factors of carbon reserves in the middle reaches of the Yangtze River urban agglomeration is crucial for achieving sustainable development and regional ecological conservation against the backdrop of the “double carbon” plan. Based on three periods of land use data from 2000 to 2020, combined with the InVEST model(Version 3.14.2), the spatiotemporal changes in carbon storage in the urban agglomeration in the middle reaches of the Yangtze River were analyzed. The PLUS model (Version 1.3.5) was used to predict three scenarios of natural development, urban development, and eco-development in the urban agglomeration in the middle reaches of the Yangtze River in 2035 and estimate the carbon storage of the ecosystems under different scenarios, and it used optimal parameter GeoDetectors (Version 4.4.2) to reveal the driving factors affecting the spatiotemporal differentiation of carbon storage. The results show that farmland and construction land area increased and forestland area continued to decrease from 2000 to 2020. Carbon storage decreased by 1 × 10⁶ t, with forestland conversion to farmland and construction land being the main decreasing drivers. The carbon storage of natural and urban developments decreased by 0.26 × 10⁶ t and 0.32 × 10⁶ t, while it increased by 0.16 × 10⁶ under ecological development. The results of the factor detector showed that the NDVI (Normalized Difference Vegetation Index) had the highest explanatory power on the spatiotemporal variation in carbon storage (q = 0.588), followed by the slope (q = 0.454) and elevation (q = 0.391), and the explanatory power of natural environmental factors on the spatiotemporal variation in of carbon storage was dominant. The interaction detector results showed that the spatiotemporal variation in carbon storage was affected by multiple factors, the interaction intensity between each driving factor was stronger than that of a single factor, and the synergy between the NDVI and slope was the strongest, at q = 0.646. Full article

Digital disparities remain a significant challenge in Indonesia, particularly across its diverse regions, with uneven access to digital infrastructure, skills, and economic opportunities. This study aims to map these digital disparities at the district level, analyze the spatial distribution and clustering of digital transformation using the Digital Society Index of Indonesia (IMDI), and investigate the key drivers of digital inequality across four core pillars: Infrastructure and Ecosystem, Digital Skills, Empowerment, and Jobs. To measure the IMDI, primary data were collected from the industrial sector and the general population over three years (2022–2024), combined with secondary data on internet usage and service standards. A multistage random sampling approach ensured representativeness, considering demographic variations and industrial segments. The analysis employed spatiotemporal methods to capture temporal trends and spatial clustering. The results revealed a significant IMDI increase from 37.80 in 2022 to 43.18 in 2023, followed by stability at 43.34 in 2024. The hotspots of digital transformation remain concentrated on Java Island, while low spots persist in eastern Indonesia. This study provides critical insights into Indonesia’s digital readiness, identifying priority areas for targeted interventions to bridge the digital divide and foster equitable digital development. Full article