Search Results (24)

Leaf litter traits among tree species exert a direct influence on spatiotemporal nutrient turnover and an indirect influence by shifting the decomposition dynamics of leaf litter mixtures including other sympatric species. Cycas micronesica and Serianthes nelsonii are two Mariana Island tree species that are endangered, and developing a greater understanding of the influence of these trees on biogeochemistry may improve information-based conservation decisions. The objectives of this study were to quantify the influence of mixing the leaf litter of these species with 12 sympatric forest plants to determine the additive and nonadditive influences on decomposition. The C. micronesica litter was collectively antagonistic when litter mixtures were incubated in a mesocosm study and a field litterbag study, and the response was similar among the included species. The S. nelsonii litter was collectively synergistic among the same mixed species, and the response was dissimilar among the included species. The contributions of these two threatened tree species to spatiotemporal diversity in biogeochemistry are dissimilar and considerable. These findings indicate that species recovery efforts for these two species are of paramount importance for maintaining Mariana Island ecological integrity and native biodiversity by sustaining their contributions to ecosystem services. Full article

(1) Understanding and characterizing the spatial and temporal variability of agricultural data is a key aspect of precision agriculture, particularly in soil management. Modeling the spatiotemporal dependency structure through geostatistical methods is essential for accurately estimating the parameters that define this structure and for performing Kriging-based interpolation. This study aimed to analyze the spatiotemporal variability of the soybean yield over ten crop years (2012–2013 to 2021–2022) in an agricultural area located in Cascavel, Paraná, Brazil. (2) Spatial analyses were conducted using two approaches: the Gaussian linear spatial model with independent multiple repetitions and the spatiotemporal model with a separable covariance structure. (3) The results showed that the maps generated using the Gaussian linear spatial model with multiple independent repetitions exhibited similar patterns to the individual soybean yield maps for each crop year. However, when comparing the kriged soybean yield maps based on independent multiple repetitions with those derived from the spatiotemporal model with a separable covariance structure, the accuracy indices indicated that the maps were dissimilar. (4) This suggests that incorporating the spatiotemporal structure provides additional information, making it a more comprehensive approach for analyzing soybean yield variability. The best model was chosen through cross-validation and a trace. Thus, incorporating a spatiotemporal model with a separable covariance structure increases the accuracy and interpretability of soybean yield analyses, making it a more effective tool for decision-making in precision agriculture. Full article

Existing infrared small target detection methods often fail due to limited exploitation of spatiotemporal information, leading to missed detections and false alarms. To address these limitations, we propose a novel framework called Spatial–Temporal Fusion Detection (STFD), which synergistically integrates three original components: gradient-enhanced spatial contrast, adaptive temporal intensity accumulation, and temporal regional contrast. In the temporal domain, we introduce Temporal Regional Contrast (TRC), the first method to quantify target-background dissimilarity through adaptive region-based temporal profiling, overcoming the rigidity of conventional motion-based detection. Concurrently, Regional Intensity Accumulation (RIA) uniquely accumulates weak target signatures across frames while suppressing transient noise, addressing the critical gap in detecting sub-SNR-threshold targets that existing temporal filters fail to resolve. For spatial analysis, we propose the Gradient-Enhanced Local Contrast Measure (GELCM), which innovatively incorporates gradient direction and magnitude coherence into contrast computation, significantly reducing edge-induced false alarms compared with traditional local contrast methods. The proposed TRC, RIA, and GELCM modules complement each other, enabling robust detection through their synergistic interactions. Specifically, our method achieves significant improvements in key metrics: SCRG increases by up to 36.59, BSF improves by up to 9.48, and AUC rises by up to 0.027, reaching over 0.99, compared with the best existing methods, indicating a substantial enhancement in detection effectiveness. Full article

A near-real-time change detection network can consistently identify unauthorized construction activities over a wide area, empowering authorities to enforce regulations efficiently. Furthermore, it can promptly assess building damage, enabling expedited rescue efforts. The extensive adoption of deep learning in change detection has prompted a predominant emphasis on enhancing detection performance, primarily through the expansion of the depth and width of networks, overlooking considerations regarding inference time and computational cost. To accurately represent the spatio-temporal semantic correlations between pre-change and post-change images, we create an innovative transformer attention mechanism named Sliding-Window Dissimilarity Cross-Attention (SWDCA), which detects spatio-temporal semantic discrepancies by explicitly modeling the dissimilarity of bi-temporal tokens, departing from the mono-temporal similarity attention typically used in conventional transformers. In order to fulfill the near-real-time requirement, SWDCA employs a sliding-window scheme to limit the range of the cross-attention mechanism within a predetermined window/dilated window size. This approach not only excludes distant and irrelevant information but also reduces computational cost. Furthermore, we develop a lightweight Siamese backbone for extracting building and environmental features. Subsequently, we integrate an SWDCA module into this backbone, forming an efficient change detection network. Quantitative evaluations and visual analyses of thorough experiments verify that our method achieves top-tier accuracy on two building change detection datasets of remote sensing imagery, while also achieving a real-time inference speed of 33.2 FPS on a mobile GPU. Full article

Background: Although computational models are advancing air quality prediction, achieving the desired performance or accuracy of prediction remains a gap, which impacts the implementation of machine learning (ML) air quality prediction models. Several models have been employed and some hybridized to enhance air quality and air quality index predictions. The objective of this paper is to systematically review machine and deep learning techniques for spatiotemporal air prediction challenges. Methods: In this review, a methodological framework based on PRISMA flow was utilized in which the initial search terms were defined to guide the literature search strategy in online data sources (Scopus and Google Scholar). The inclusion criteria are articles published in the English language, document type (articles and conference papers), and source type (journal and conference proceedings). The exclusion criteria are book series and books. The authors’ search strategy was complemented with ChatGPT-generated keywords to reduce the risk of bias. Report synthesis was achieved by keyword grouping using Microsoft Excel, leading to keyword sorting in ascending order for easy identification of similar and dissimilar keywords. Three independent researchers were used in this research to avoid bias in data collection and synthesis. Articles were retrieved on 27 July 2024. Results: Out of 374 articles, 80 were selected as they were in line with the scope of the study. The review identified the combination of a machine learning technique and deep learning techniques for data limitations and processing of the nonlinear characteristics of air pollutants. ML models, such as random forest, and decision tree classifier were among the commonly used models for air quality index and air quality predictions, with promising performance results. Deep learning models are promising due to the hyper-parameter components, which consist of activation functions suitable for nonlinear spatiotemporal data. The emergence of low-cost devices for data limitations is highlighted, in addition to the use of transfer learning and federated learning models. Again, it is highlighted that military activities and fires impact the O₃ concentration, and the best-performing models highlighted in this review could be helpful in developing predictive models for air quality prediction in areas with heavy military activities. Limitation: This review acknowledges methodological challenges in terms of data collection sources, as there are equally relevant materials on other online data sources. Again, the choice and use of keywords for the initial search and the creation of subsequent filter keywords limit the collection of other relevant research articles. Full article

The Sicilian wetlands (Italy) are seriously threatened by human activities and ongoing climate change. The loss of these wetlands as migratory stepping stones could severely hamper the migratory flow of many bird species along the central Mediterranean. Targeted actions for the conservation of the avifauna require thorough knowledge of the utilization that waterbirds make of these habitats. Aiming to inform planning for more satisfactory bird habitat management and bird diversity preservation along the Mediterranean migratory bird flyway, in this study, we inventoried the avian metacommunity of the coastal wetlands in Sicily during the most critical period of the year (July–September) and used a multilevel analytical framework to explore the spatiotemporal dynamics in bird species assemblages. We recorded 73 bird species, of which almost 90% were migratory and 30 belonged to Annex I of the Birds Directive. At the metacommunity level, we found that all the biodiversity metrics were low in July and approximately doubled in the successive sampling sessions (August–September), where they showed little if any change. At the community level, we detected two main clusters of wetlands with regard to species richness, of which one (wetlands Baronello, Gela, Gornalunga, and Roveto) was characterized by higher levels of species richness in nearly all the sampling dates. The pattern of species richness in the Sicilian wetlands was most similar between the first and second half of August, while July was very dissimilar from all the other sampling dates. At the guild level, we found a significant increase during July–September in the number of the species belonging to the “Mediterranean” migration guild and the “divers from the surface” and “surface feeders” foraging guilds. At the species level, we detected a significant temporal sequence of the occurrence of waterbird species: two species were only early dwellers in July, ten species were only late dwellers in September, and twenty-six species made use of the Sicilian wetlands all summer long. The spatial distribution of the waterbird species differed significantly between any pair of sampling dates. Overall, the Little Grebe, the Spotted Redshank, and the Little Tern were the bird species with the highest site infidelity; by contrast, the Black Stork, the Broad-billed Sandpiper, the European Golden Plover, the Common Shelduck, and the Black-necked Grebe changed their spatial distribution among wetlands the least during July–September. Our study allowed us to detect (1) the wetlands and (2) the waterbird species to which the priority for conservation should be assigned, as well as (3) the exact time span during July–September when conservation measures should be mandatory, and not only advisable. These results provide a broader insight of the space–time patterns in bird species assembly in the coastal wetlands of Sicily during the critical summer period. Full article

Burst events in Water Distribution Networks (WDNs) pose a significant threat to the safety of water supply, leading people to focus on efficient methods for burst localization and prompt repair. This paper proposes a multi-stage burst localization method, which includes preliminary region determination and precise localization analysis. Based on the hydraulic model and spatio-temporal information, the effective sensor sequences and monitoring areas of the nodes are determined. In the first stage, the preliminary burst region is determined based on the monitoring region of sensors and the alarm sensors. In the second stage, localization metrics are used to analyze the dissimilarity degree between burst data from the hydraulic model and the monitoring data from the effective sensors at each node. This analysis helps identify candidate burst nodes and determine their localization priorities. The localization model is tested on the C-Town network to obtain comparative results. The method effectively reduces the burst region, minimizes the search region, and significantly improves the efficiency of burst localization. For precise localization, it accurately localizes the burst event by prioritizing the possibilities of the burst location. Full article

Image-based gender classification is very useful in many applications, such as intelligent surveillance, micromarketing, etc. One common approach is to adopt a machine learning algorithm to recognize the gender class of the captured subject based on spatio-temporal gait features extracted from the image. The image input can be generated from the video of the walking cycle, e.g., gait energy image (GEI). Recognition accuracy depends on the similarity of intra-class GEIs, as well as the dissimilarity of inter-class GEIs. However, we observe that, at some viewing angles, the GEIs of both gender classes are very similar. Moreover, the GEI does not exhibit a clear appearance of posture. We postulate that distinctive postures of the walking cycle can provide additional and valuable information for gender classification. This paper proposes a gender classification framework that exploits multiple inputs of the GEI and the characteristic poses of the walking cycle. The proposed framework is a cascade network that is capable of gradually learning the gait features from images acquired in multiple views. The cascade network contains a feature extractor and gender classifier. The multi-stream feature extractor network is trained to extract features from the multiple input images. Features are then fed to the classifier network, which is trained with ensemble learning. We evaluate and compare the performance of our proposed framework with state-of-the-art gait-based gender classification methods on benchmark datasets. The proposed framework outperforms other methods that only utilize a single input of the GEI or pose. Full article

Machine learning-based gait systems facilitate the real-time control of gait assistive technologies in neurological conditions. Improving such systems needs the identification of kinematic signals from inertial measurement unit wearables (IMUs) that are robust across different walking conditions without extensive data processing. We quantify changes in two kinematic signals, acceleration and angular velocity, from IMUs worn on the frontal plane of bilateral shanks and thighs in 30 adolescents (8–18 years) on a treadmills and outdoor overground walking at three different speeds (self-selected, slow, and fast). Primary curve-based analyses included similarity analyses such as cosine, Euclidean distance, Poincare analysis, and a newly defined bilateral symmetry dissimilarity test (BSDT). Analysis indicated that superior–inferior shank acceleration (SI shank Acc) and medial–lateral shank angular velocity (ML shank AV) demonstrated no differences to the control signal in BSDT, indicating the least variability across the different walking conditions. Both SI shank Acc and ML shank AV were also robust in Poincare analysis. Secondary parameter-based similarity analyses with conventional spatiotemporal gait parameters were also performed. This normative dataset of walking reports raw signal kinematics that demonstrate the least to most variability in switching between treadmill and outdoor walking to help guide future machine learning models to assist gait in pediatric neurological conditions. Full article

Weather conditions at the time of wildfire front arrival strongly influence fire behavior and effects, yet few methods exist for estimating weather conditions more spatio-temporally resolved than coarse-grain (e.g., 4 km) daily averages. When a fire front advances rapidly and weather conditions are heterogeneous over space and time, greater spatio-temporal precision is required to accurately link fire weather to observed fire effects. To identify the influence of fire weather on fire effects observed across a sample of existing forest inventory plots during a wind-driven megafire event in the US Pacific Northwest, we explored and compared three methods for estimating time of fire arrival and the wind speed at that arrival time for each plot location. Two methods were based on widely used, remotely sensed active fire data products with dissimilar spatial and temporal resolutions. The third and preferred method, Modeled-Weather Interpolated Perimeters (MoWIP), is a new approach that leveraged fine-grained (1.3 km, hourly) wind speed and direction from modeled fire weather to guide interpolation of aerial infrared-detected (IR) operational perimeters, subdividing the time intervals defined by sequential IR perimeters into quartile intervals to enhance temporal resolution of predicted fire arrival times. Our description of these fire arrival “time stamp” methods and discussion of their utility and shortcomings should prove useful to fire scientists, ecologists, land managers, and future analyses seeking to link estimated fire weather and observed fire effects. Full article

Both the vastness of pasturelands and the value they contain—e.g., food security, ecosystem services—have resulted in increased scientific and industry efforts to remotely monitor them via satellite imagery and machine learning (ML). However, the transferability of these models is uncertain, as modelers commonly train and test on site-specific or homogenized—i.e., randomly partitioned—datasets and choose complex ML algorithms with increased potential to overfit a limited dataset. In this study, we evaluated the accuracy and transferability of remote sensing pasture models, using multiple ML algorithms and evaluation structures. Specifically, we predicted pasture above-ground biomass and nitrogen concentration from Sentinel-2 imagery. The implemented ML algorithms include principal components regression (PCR), partial least squares regression (PLSR), least absolute shrinkage and selection operator (LASSO), random forest (RF), support vector machine regression (SVR), and a gradient boosting model (GBM). The evaluation structures were determined using levels of spatial and temporal dissimilarity to partition the train and test datasets. Our results demonstrated a general decline in accuracy as evaluation structures increase in spatiotemporal dissimilarity. In addition, the more simplistic algorithms—PCR, PLSR, and LASSO—out-performed the more complex models RF, SVR, and GBM for the prediction of dissimilar evaluation structures. We conclude that multi-spectral satellite and pasture physiological variable datasets, such as the one presented in this study, contain spatiotemporal internal dependence, which makes the generalization of predictive models to new localities challenging, especially for complex ML algorithms. Further studies on this topic should include the assessment of model transferability by using dissimilar evaluation structures, and we expect generalization to improve for larger and denser datasets. Full article

Pheromone-baited traps have been widely used for the monitoring of the cotton bollworm, Helicoverpa armigera (Hübner) (Lepidoptera: Noctuidae), in order to time any control measures during the growing season. Different monitoring techniques may provide differential results regarding adult captures. However, studies on the comparative evaluation of the performance of different trap types on the captures of H. armigera are limited. To close this gap, in the present study, three different funnel traps (striped, green, and colored) were simultaneously evaluated in Central Greece, one of the main cotton-producing geographical zones in the European Union, in order to compare trap performance on the captures of H. armigera, as well as to depict the distribution of this species per trap in the study area. A differential performance of the different trap types tested, expressed as numbers of adults captured, was recorded. Specifically, the striped trap captured many more adult moths than the other two trap types. Given that the only difference among these traps was the color of the external trap surface, we hypothesize that trap color does matter in the case of H. armigera, and it is likely that brighter colors may be more attractive than darker ones. Full article

With the continuous development of smart cities, intelligent transportation systems (ITSs) have ushered in many breakthroughs and upgrades. As a solid foundation for an ITS, traffic flow prediction effectively helps the city to better manage intricate traffic flow. However, existing traffic flow prediction methods such as temporal graph convolutional networks(T-GCNs) ignore the dissimilarities between lanes. Thus, they cannot provide more specific information regarding predictions such as dynamic changes in traffic flow direction and deeper lane relationships. With the upgrading of intersection sensors, more and more intersection lanes are equipped with intersection sensors to detect vehicle information all day long. These spatio-temporal data help researchers refine the focus of traffic prediction research down to the lane level. More accurate and detailed data mean that it is more difficult to mine the spatio-temporal correlations between data, and modeling heterogeneous data becomes more challenging. In order to deal with these problems, we propose a heterogeneous graph convolution model based on dynamic graph generation. The model consists of three components. The internal graph convolution network captures the real-time spatial dependency between lanes in terms of generated dynamic graphs. The external heterogeneous data fusion network comprehensively considers other parameters such as lane speed, lane occupancy, and weather conditions. The codec neural network utilizes a temporal attention mechanism to capture the deep temporal dependency. We test the performance of this model based on two real-world datasets, and extensive comparative experiments indicate that the proposed heterogeneous graph convolution model can improve the prediction accuracy. Full article

Outbreaks of SARS-CoV-2 can be attributed to expanding small-scale localized infection subclusters that eventually propagate into regional and global outspread. These infections are driven by spatial as well as temporal mutational dynamics wherein virions diverge genetically as transmission occurs. Mutational similarity or dissimilarity of viral strains, stemming from shared spatiotemporal fields, thence serves as a gauge of relatedness. In our clinical laboratory, molecular epidemiological analyses of strain association are performed qualitatively from genomic sequencing data. These methods however carry a degree of uncertainty when the samples are not qualitatively, with reasonable confidence, deemed identical or dissimilar. We propose a theoretical mathematical model for probability derivation of outbreak-sample similarity as a function of spatial dynamics, shared and different mutations, and total number of samples involved. This Similarity Index utilizes an Essen-Möller ratio of similar and dissimilar mutations between the strains in question. The indices are compared to each strain within an outbreak, and then the final Similarity Index of the outbreak group is calculated to determine quantitative confidence of group relatedness. We anticipate that this model will be useful in evaluating strain associations in SARS-CoV-2 and other viral outbreaks utilizing molecular data. Full article

Protected areas (PAs) transform over time due to natural and anthropogenic processes, resulting in the loss of biodiversity and ecosystem services. As current and projected climatic trends are poised to pressurize the sustainability of PAs, analyses of the existing perturbations are crucial for providing valuable insights that will facilitate conservation management. In this study, land cover change, landscape characteristics, and spatiotemporal patterns of the vegetation intensity in the Kasungu National Park (area = 2445.10 km²) in Malawi were assessed using Landsat data (1997, 2008 and 2018) in a Fuzzy K-Means unsupervised classification. The findings reveal that a 21.12% forest cover loss occurred from 1997 to 2018: an average annual loss of 1.09%. Transition analyses of the land cover changes revealed that forest to shrubs conversion was the main form of land cover transition, while conversions from shrubs (3.51%) and bare land (3.48%) to forest over the two decades were comparatively lower, signifying a very low rate of forest regeneration. The remaining forest cover in the park was aggregated in a small land area with dissimilar landscape characteristics. Vegetation intensity and vigor were lower mainly in the eastern part of the park in 2018. The findings have implications for conservation management in the context of climate change and the growing demand for ecosystem services in forest-dependent localities. Full article