Large-scale Variations Research Articles

Atmospheric Trace Gas Oxidizers Contribute to Soil Carbon Fixation Driven by Key Soil Conditions in Terrestrial Ecosystems.

Microbial oxidizers of trace gases such as hydrogen (H2) and carbon monoxide (CO) are widely distributed in soil microbial communities and play a vital role in modulating biogeochemical cycles. However, the contribution of trace gas oxidizers to soil carbon fixation and the driving environmental factors remain unclear, especially on large scales. Here, we utilized biogeochemical and genome-resolved metagenomic profiling, assisted by machine learning analysis, to estimate the contributions of trace gas oxidizers to soil carbon fixation and to predict the key environmental factors driving this process in soils from five distinct ecosystems. The results showed that phylogenetically and physiologically diverse H2 and CO oxidizers and chemosynthetic carbon-fixing microbes are present in the soil in different terrestrial ecosystems. The large-scale variations in soil carbon fixation were highly positively correlated with both the abundance and the activity of H2 and CO oxidizers (p < 0.05-0.001). Furthermore, soil pH and moisture-induced shifts in the abundance of H2 and CO oxidizers partially explained the variation in soil carbon fixation (55%). The contributions of trace gas oxidizers to soil carbon fixation in the different terrestrial ecosystems were estimated to range from 1.1% to 35.0%. The estimated rate of trace gas carbon fixation varied from 0.04 to 1.56 mg kg-1 d-1. These findings reveal that atmospheric trace gas oxidizers may contribute to soil carbon fixation driven by key soil environmental factors, highlighting the non-negligible contribution of these microbes to terrestrial carbon cycling.

High-quality genome of black wolfberry (Lycium ruthenicum Murr.) provides insights into the genetics of anthocyanin biosynthesis regulation

Abstract Black wolfberry (Lycium ruthenicum Murr.) is an important plant for ecological preservation. In addition, its fruits are rich in anthocyanins and have important edible and medicinal value. However, a high quality chromosome-level genome for this species is not yet available, and the regulatory mechanisms involved in the biosynthesis of anthocyanins are unclear. In this study, haploid material was used to assemble a high-quality chromosome-level reference genome of Lycium ruthenicum, resulting in a genome size of 2,272 Mb with contig N50 of 92.64 Mb, and 38,993 annotated gene models. In addition, the evolution of this genome and large-scale variations compared with the Ningxia wolfberry Lycium barbarum were determined. Importantly, homology annotation identified 86 genes involved in the regulatory pathway of anthocyanin biosynthesis, five of which [LrCHS1 (evm.TU.Chr05.295), LrCHS2 (evm.TU.Chr09.488), LrAOMT (evm.TU.Chr09.809), LrF3'5'H (evm.TU.Chr06.177) and LrAN2.1 (evm.TU.Chr05.2618)] were screened by differential expression analysis and correlation analysis using a combination of transcriptome and metabolome testing. Overexpression of these genes could significantly up- or downregulate anthocyanin-related metabolites. These results will help accelerate the functional genomic research of L. ruthenicum, and the elucidation of the genes involved in anthocyanin synthesis will be beneficial for breeding new varieties and further exploring its ecological conservation potential.

Semantic Segmentation of Satellite Images for Landslide Detection Using Foreground-Aware and Multi-Scale Convolutional Attention Mechanism.

Advancements in satellite and aerial imagery technology have made it easier to obtain high-resolution remote sensing images, leading to widespread research and applications in various fields. Remote sensing image semantic segmentation is a crucial task that provides semantic and localization information for target objects. In addition to the large-scale variation issues common in most semantic segmentation datasets, aerial images present unique challenges, including high background complexity and imbalanced foreground-background ratios. However, general semantic segmentation methods primarily address scale variations in natural scenes and often neglect the specific challenges in remote sensing images, such as inadequate foreground modeling. In this paper, we present a foreground-aware remote sensing semantic segmentation model. The model introduces a multi-scale convolutional attention mechanism and utilizes a feature pyramid network architecture to extract multi-scale features, addressing the multi-scale problem. Additionally, we introduce a Foreground-Scene Relation Module to mitigate false alarms. The model enhances the foreground features by modeling the relationship between the foreground and the scene. In the loss function, a Soft Focal Loss is employed to focus on foreground samples during training, alleviating the foreground-background imbalance issue. Experimental results indicate that our proposed method outperforms current state-of-the-art general semantic segmentation methods and transformer-based methods on the LS dataset benchmark.

MCH-PAN: gastrointestinal polyp detection model integrating multi-scale feature information.

The rise of object detection models has brought new breakthroughs to the development of clinical decision support systems. However, in the field of gastrointestinal polyp detection, there are still challenges such as uncertainty in polyp identification and inadequate coping with polyp scale variations. To address these challenges, this paper proposes a novel gastrointestinal polyp object detection model. The model can automatically identify polyp regions in gastrointestinal images and accurately label them. In terms of design, the model integrates multi-channel information to enhance the ability and robustness of channel feature expression, thus better coping with the complexity of polyp structures. At the same time, a hierarchical structure is constructed in the model to enhance the model's adaptability to multi-scale targets, effectively addressing the problem of large-scale variations in polyps. Furthermore, a channel attention mechanism is designed in the model to improve the accuracy of target positioning and reduce uncertainty in diagnosis. By integrating these strategies, the proposed gastrointestinal polyp object detection model can achieve accurate polyp detection, providing clinicians with reliable and valuable references. Experimental results show that the model exhibits superior performance in gastrointestinal polyp detection, which helps improve the diagnostic level of digestive system diseases and provides useful references for related research fields.

ALMA-IMF

The stellar initial mass function (IMF) is critical to our understanding of star formation and the effects of young stars on their environment. On large scales, it enables us to use tracers such as UV or Hα emission to estimate the star formation rate of a system and interpret unresolved star clusters across the Universe. So far, there is little firm evidence of large-scale variations of the IMF, which is thus generally considered “universal”. Stars form from cores, and it is now possible to estimate core masses and compare the core mass function (CMF) with the IMF, which it presumably produces. The goal of the ALMA-IMF large programme is to measure the core mass function at high linear resolution (2700 au) in 15 typical Milky Way protoclusters spanning a mass range of 2.5 × 103 to 32.7 × 103 M⊙. In this work, we used two different core extraction algorithms to extract ≈680 gravitationally bound cores from these 15 protoclusters. We adopted a per core temperature using the temperature estimate from the point-process mapping Bayesian method (PPMAP). A power-law fit to the CMF of the sub-sample of cores above the 1.64 M⊙ completeness limit (330 cores) through the maximum likelihood estimate technique yields a slope of 1.97 ± 0.06, which is significantly flatter than the 2.35 Salpeter slope. Assuming a self-similar mapping between the CMF and the IMF, this result implies that these 15 high-mass protoclusters will generate atypical IMFs. This sample currently is the largest sample that was produced and analysed self-consistently, derived at matched physical resolution, with per core temperature estimates, and cores as massive as 150 M⊙. We provide both the raw source extraction catalogues and the catalogues listing the source size, temperature, mass, spectral indices, and so on in the 15 protoclusters.

A unified feature-motion consistency framework for robust image matching

Establishing reliable feature matches between a pair of images in various scenarios is a long-standing open problem in photogrammetry. Attention-based detector-free matching with coarse-to-fine architecture has been a typical pipeline to build matches, but the cross-attention module with global receptive field may compromise the structural local consistency by introducing irrelevant regions (outliers). Motion field can maintain structural local consistency under the assumption that matches for adjacent features should be spatially proximate. However, motion field can only estimate local displacements between consecutive images and struggle with long-range displacements estimation in large-scale variation scenarios without spatial correlation priors. Moreover, large-scale variations may also disrupt the geometric consistency with the application of mutual nearest neighbor criterion in patch-level matching, making it difficult to recover accurate matches. In this paper, we propose a unified feature-motion consistency framework for robust image matching (MOMA), to maintain structural consistency at both global and local granularity in scale-discrepancy scenarios. MOMA devises a motion consistency-guided dependency range strategy (MDR) in cross attention, aggregating highly relevant regions within the motion consensus-restricted neighborhood to favor true matchable regions. Meanwhile, a unified framework with hierarchical attention structure is established to couple local motion field with global feature correspondence. The motion field provides local consistency constraints in feature aggregation, while feature correspondence provides spatial context prior to improve motion field estimation. To alleviate geometric inconsistency caused by hard nearest neighbor criterion, we propose an adaptive neighbor search (soft) strategy to address scale discrepancy. Extensive experiments on three datasets demonstrate that our method outperforms solid baselines, with AUC improvements of 4.73/4.02/3.34 in two-view pose estimation task at thresholds of 5°/10°/20° on Megadepth test, and 5.94% increase of accuracy at threshold of 1px in homography task on HPatches datasets. Furthermore, in the downstream tasks such as 3D mapping, the 3D models reconstructed using our method on the self-collected SYSU UAV datasets exhibit significant improvement in structural completeness and detail richness, manifesting its high applicability in wide downstream tasks. The code is publicly available at https://github.com/BunnyanChou/MOMA.

Structure of Reef Fish Families (Butterflyfishes and Angelfishes) at Isolated Oceanic Reefs in the Indian Ocean: Christmas Island and the Cocos (Keeling) Islands

There has been substantial research on the factors that structure reef fish assemblages, but this has mostly focused on nearshore or continental reefs. This study examines patterns of abundance and species composition for two iconic groups of coral reef fishes, angelfishes (family Pomacanthidae) and butterflyfishes (family Chaetodontidae) at two isolated, oceanic reefs in the Indian Ocean: Christmas Island and the Cocos (Keeling) Islands. Six explanatory variables were investigated to determine whether large-scale physical factors are more important than fine-scale biotic factors in structuring reef fish communities on oceanic islands. For angelfishes, depth was the factor that most explained patterns in abundance (explaining 46.5% of the variation), species richness (44.8%) and composition (15.3%), with both abundance and species richness being greater at 20 m than at 5 m. Differences in species composition were greater between depths than between islands. For butterflyfishes, variation in abundance and species richness was best explained by the difference in aspect or exposure among sites, though abundance and composition also differed significantly between Christmas Island and the Cocos (Keeling) Islands. Large-scale variation in the structure of these reef fish assemblages could not be explained based on differences in habitat diversity, or coral and algal cover. This study indicates that large-scale physical factors (island location, exposure, depth) are more important than fine-scale biotic factors in structuring reef fish assemblages on oceanic islands.

The influence of topography on subaqueous gravity flows: a case study from the Jurassic Los Molles Formation, Neuquén Basin, Argentina

ABSTRACT Subaqueous sediment gravity flows experience modifications when they interact with slope and basin-floor topography, impacting facies, geometries, and architectural patterns of the deposits. Understanding these processes is critical for reservoir-quality and trap predictions in deep-water stratigraphic plays. A comprehensive literature summary of the topographic influence on subaqueous gravity flows and an interpretation applied to the basin-floor turbidite system of the late synrift Los Molles Formation in the Neuquén Basin of Argentina are provided. In the study area, a high-resolution satellite image, drone imagery, and 30 logs (about 5000 m total thickness) were measured, focusing on facies analysis, paleocurrents, bed thicknesses, and large-scale thickness variations. Studies describing topographic confinement of sediment gravity flows have used five approaches: 1) Paleocurrent analysis: paleocurrents following structural trends and variations within the same bed (flow deflection and reflection). 2) Facies analysis: complex facies variations on short distances and increased number of hybrid beds and debrites near barriers (flow transformation), loading, convoluted structures, and bidirectional ripples (flow reflection), and sand–mud couplets or thick mudcaps (flow ponding). 3) Small-scale (decimeter to meter) thickness variations: beds thickening towards topography, beds thinning on top of topography (onlaps), and low thinning rates (loss of flow competence, flow containment). 4) Architectural analysis: aggradational stacking, lateral stacking away from topographic barriers, and fill–spill successions. 5) Large-scale thickness (tens of meters) variations: increased fan system thickness across topographic lows. An extended version of an existing classification scheme on degree of confinement is proposed: C0 (unconfined): no flow modification evidence present, and compensational stacking is a classic architectural pattern. C1 (weakly confined): variations in regional paleocurrent directions, minor thickness variations against small-relief barriers, large (basin scale) thickness variations. C2a (confined) to C2b (highly confined): onlaps against high-relief barriers, paleocurrent direction variations on the same bed, bed thickening against topography, facies evidencing flow reflection, increased number of hybrid beds, aggradational stacking, and high percentage of beds continuous over 500 m distances (tabular beds). C2a and C2b are considered end members of a spectrum where the amount of evidence listed can vary. C3 (ponded): sand mud couplets or thick mudcaps. Higher confinement categories might have indicators from the weaker confinement categories. The lower fan in the Los Molles Fm. turbidite system is classified as weakly confined. A complex sediment routing followed structural trends inherited from the rift phase of the basin. The basin-floor had irregular water depths that led to preferred sites of deposition. Hybrid beds are mostly found at the fringes of the lower part of the succession and might be associated with an above-grade slope profile on a footwall scarp. Lastly, the Los Molles Fm. basin-floor beds present high thinning rates (average 1.5 m/km) and the system stacks compensationally, pointing to null interaction against basin margins.

Transformer fusion for indoor RGB-D semantic segmentation

Fusing geometric cues with visual appearance is an imperative theme for RGB-D indoor semantic segmentation. Existing methods commonly adopt convolutional modules to aggregate multi-modal features, paying little attention to explicitly leveraging the long-range dependencies in feature fusion. Therefore, it is challenging for existing methods to accurately segment objects with large-scale variations. In this paper, we propose a novel transformer-based fusion scheme, named TransD-Fusion, to better model contextualized awareness. Specifically, TransD-Fusion consists of a self-refinement module, a calibration scheme with cross-interaction, and a depth-guided fusion. The objective is to first improve modality-specific features with self- and cross-attention, and then explore the geometric cues to better segment objects sharing a similar visual appearance. Additionally, our transformer fusion benefits from a semantic-aware position encoding which spatially constrains the attention to neighboring pixels. Extensive experiments on RGB-D benchmarks demonstrate that the proposed method performs well over the state-of-the-art methods by large margins.

MMCA-NET: A Multimodal Cross Attention Transformer Network for Nasopharyngeal Carcinoma Tumor Segmentation Based on a Total-Body PET/CT System.

Nasopharyngeal carcinoma (NPC) is a malignant tumor primarily treated by radiotherapy. Accurate delineation of the target tumor is essential for improving the effectiveness of radiotherapy. However, the segmentation performance of current models is unsatisfactory due to poor boundaries, large-scale tumor volume variation, and the labor-intensive nature of manual delineation for radiotherapy. In this paper, MMCA-Net, a novel segmentation network for NPC using PET/CT images that incorporates an innovative multimodal cross attention transformer (MCA-Transformer) and a modified U-Net architecture, is introduced to enhance modal fusion by leveraging cross-attention mechanisms between CT and PET data. Our method, tested against ten algorithms via fivefold cross-validation on samples from Sun Yat-sen University Cancer Center and the public HECKTOR dataset, consistently topped all four evaluation metrics with average Dice similarity coefficients of 0.815 and 0.7944, respectively. Furthermore, ablation experiments were conducted to demonstrate the superiority of our method over multiple baseline and variant techniques. The proposed method has promising potential for application in other tasks.

Employing scientific evidence through synthetic indices to support management and conservation policies for an endangered tree of the Gran Chaco

Management and conservation programs require accessible scientific evidence to effectively plan and achieve their goals. Facing regional and species-specific threats, Gonopterodendron sarmientoi, an endangered tree native to the Gran Chaco region, requires scientific evidence to support management and conservation actions, notably in forest management and territorial planning. To bridge this crucial gap, we developed indices that integrate genetic, evolutionary, morphological, threat-related, and ethnobotanical data concerning G. sarmientoi at a landscape level. Across twenty-four localities of this species, we have created four distinct indices. The first index amalgamates data on genetics, functional traits, and paleoclimate information. The second index incorporates data on threats stemming from deforestation and climate change scenarios. The third index provides supplementary insights into the species’ utilization by indigenous groups. Finally, the fourth index combines all the aforementioned data to prioritize the most valuable and threatened localities. Through this comprehensive approach, we have identified specific in-situ and ex-situ management and conservation actions for G. sarmientoi. This approach considers broader-scale conservation efforts and considers large-scale variations and processes. The indices effectively link scientific data with actionable insights, providing decision-makers involved with G. sarmientoi with concise information that encapsulates the most pertinent aspects of species conservation.

Seasonal and environmental factors contribute to the variation in the gut microbiome: A large-scale study of a small bird.

Environmental variation can shape the gut microbiome, but broad/large-scale data on among and within-population heterogeneity in the gut microbiome and the associated environmental factors of wild populations is lacking. Furthermore, previous studies have limited taxonomical coverage, and knowledge about wild avian gut microbiomes is still scarce. We investigated large-scale environmental variation in the gut microbiome of wild adult great tits across the species' European distribution range. We collected fecal samples to represent the gut microbiome and used the 16S rRNA gene sequencing to characterize the bacterial gut microbiome. Our results show that gut microbiome diversity is higher during winter and thatthere are compositional differences between winter and summer gut microbiomes. During winter, individuals inhabiting mixed forest habitat show higher gut microbiome diversity, whereas there was no similar association during summer. Also, temperature was found to be a small contributor to compositional differences in the gut microbiome. We did not find significant differences in the gut microbiome among populations, nor any association between latitude, rainfall and the gut microbiome. The results suggest that there is a seasonal change in wild avian gut microbiomes, but that there are still many unknown factors that shape the gut microbiome of wild bird populations.

Surface Defect Detection for Aerospace Aluminum Profiles with Attention Mechanism and Multi-Scale Features

A YOLOv5 aluminum profile defect detection algorithm that integrates attention and multi-scale features is proposed in this paper to address the issues of the low detection accuracy, high false detection rates, and high missed detection rates that are caused by the large-scale variation of surface defects, inconspicuous small defect characteristics, and a lack of concentrated feature information in defect areas. Firstly, an improved CBAM (Channel-Wise Attention Module) convolutional attention module is employed, which effectively focuses on the feature information of defect areas in the aluminum defect dataset with only a small amount of spatial dimension. Secondly, a bidirectional weighted feature fusion network is utilized, incorporating a multi-scale feature fusion network with skip connections to aggregate various high-resolution features, thus enriching the semantic expression of features. Then, new size feature maps that have not been fused are introduced into the detection layer network to improve the detection effect of small target defects. Experimental results indicate that an average detection accuracy (mAP) of 82.6% was achieved by the improved YOLOv5 algorithm on the aluminum surface defect dataset. An improvement of 6.2% over the previous version was observed. The current defect detection requirements of aluminum profile production sites are met by this enhanced algorithm.

LHFFNet: A hybrid feature fusion method for lane detection

Lane line images have the essential attribute of large-scale variation and complex scene information, and the similarity between adjacent lane lines is high, which can easily cause classification errors. And remote lane lines are difficult to recognize due to visual angle changes in width. To address this issue, this paper proposes an effective lane detection framework, which is a hybrid feature fusion network that enhances multiple spatial features and distinguishes key features throughout the entire lane line segment. It enhances and fuses lane line features at multiscale to enhance the feature representation of lane line images, especially at the far end. Firstly, in order to enhance the correlation of multiscale lane features, a multi-head self attention is used to construct a multi-space attention enhancement module for feature enhancement in multispace. Secondly, a spatial separable convolutional branch is designed for the jumping layer structure connecting multiscale lane line features. While retaining feature information of different scales, important lane areas in multiscale feature information are emphasized through the allocation of spatial attention weights. Finally, considering that lane lines are elongated areas in the image, and the background information in the image is much more abundant than lane line information, the flexibility of traditional pooling operations in capturing widely existing anisotropic contexts in actual environments is limited. Therefore, before embedding feature output branches, strip pooling is introduced to refine the representation of lane line information and optimize model performance. The experimental results show that the accuracy on the TuSimple dataset reaches 96.84%, and the F1 score on the CULane dataset reaches 75.9%.

Efficient YOLOv8 algorithm for extreme small-scale object detection

The past decade has witnessed significant progress in detecting small objects that are often distributed with large-scale variations in arbitrary orientations. However, most of the existing models are unable to detect extremely small objects, due to (1) tiny-scale objects with more blurring in comparison to ground-view objects provide less important information for precise and robust recognition; (2) Unevenly distributed objects make detection inefficient, particularly in regions occupied by dense objects, which lead to the common inconsistency between the classification score and localization accuracy. This research proposes an efficient YOLOv8 model with a new backbone network, and anchor-free head to address these issues. Bottleneck layers are used in the backbone, allowing the model to capture more complicated patterns and reduce the dimensionality of the feature maps while maintaining the same number of channels. Whereas an anchor-free head detection module does not require pre-defined anchor boxes, they directly predict objects’ center points and sizes and their corresponding class probabilities. This proposed model is experimentally verified on the VisDrone dataset and achieved higher mAP (45.9 %), and precision (76.8 %). These experimental results suggest that the proposed model outperforms existing models in detecting x-small items.

ASD-YOLO: An aircraft surface defects detection method using deformable convolution and attention mechanism

Aircraft surface defect (ASD) detection is crucial for ensuring flight safety. Addressing challenges such as large-scale variations, irregular shapes, and sample imbalance in ASD detection, this paper proposes a ASD-YOLO network based on YOLOv5. ASD-YOLO incorporates several enhancements to improve recognition capabilities. Firstly, a new deformable convolutional feature extraction module (DCNC3) is designed to better learn defects of different shapes, which is combined with a global attention mechanism (GAM) to pay more attention to defects region information. Secondly, the feature representation of small defects is bolstered by the contextual enhancement module (CEM). Lastly, to alleviate sample imbalance problem, we introduce an exponential sliding average weight function (EMA-Slide). Experimental results on two datasets show improvements in mean Average Precision (mAP) by 5.7% and 3.4%, respectively, surpassing mainstream algorithms and offering a novel approach to ASD detection.

Genomic variation of 363 diverse tea accessions unveils the genetic diversity, domestication, and structural variations associated with tea adaptation.

Domestication has shaped the population structure and agronomic traits of tea plants, yet the complexity of tea population structure and genetic variation that determines these traits remains unclear. We here investigated the resequencing data of 363 diverse tea accessions collected extensively from almost all tea distributions and found that the population structure of tea plants was divided into eight subgroups, which were basically consistent with their geographical distributions. The genetic diversity of tea plants in China decreased from southwest to east as latitude increased. Results also indicated that Camellia sinensis var. assamica (CSA) illustrated divergent selection signatures with Camellia sinensis var. sinensis (CSS). The domesticated genes of CSA were mainly involved in leaf development, flavonoid and alkaloid biosynthesis, while the domesticated genes in CSS mainly participated in amino acid metabolism, aroma compounds biosynthesis, and cold stress. Comparative population genomics further identified ~730 Mb novel sequences, generating 6,058 full-length protein-encoding genes, significantly expanding the gene pool of tea plants. We also discovered 217,376 large-scale structural variations and 56,583 presence and absence variations (PAVs) across diverse tea accessions, some of which were associated with tea quality and stress resistance. Functional experiments demonstrated that two PAV genes (CSS0049975 and CSS0006599) were likely to drive trait diversification in cold tolerance between CSA and CSS tea plants. The overall findings not only revealed the genetic diversity and domestication of tea plants, but also underscored the vital role of structural variations in the diversification of tea plant traits.

Multi-scale keypoints detection and motion features extraction in dairy cows using ResNet101-ASPP network

Detecting Keypoints in dairy cows aims to locate and track the motion trajectories of the body's joints, which plays a crucial role in behavior analysis and lameness detection. However, real farming scenarios, characterized by occlusions and large variations in object scale may result in poor detection results. Therefore we introduce the atrous spatial pyramid pooling (ASPP) module into the shallow network of ResNet101, designed to improve the multi-scale feature extraction capability of the model. The ASPP module enhances the robustness of recognition for different dimensional sizes and occluded keypoints using different dilatation rates in the parallel atrous convolutional layers to expand the model's receptive field. Furthermore, seven types of motion features, including tracking up, gait symmetry, step height balance, motion speed variability, head swing amplitude, head-neck slope and back curvature are extracted simultaneously by monitoring and tracking the motion trajectory of distinct keypoints. Several of these features represent innovative extraction models and attributes, first proposed in this study. Multiple models are trained and tested on datasets containing 2385 frames for ablation experiments. The experiments show that, in comparison with the ResNet50, MobileNet_v2_1.0, and EfficientNet-b0 backbone networks, the training error and test error of ResNet101 improve by 4.04–30.12 pixels and 3.81–28.14 pixels. Therefore, ResNet101 is used as the benchmark for subsequent model improvement by adding the ASPP module. The training error and test error of the ResNet101-ASPP network are improved by 0.27 and 0.24 pixels, respectively, compared to the benchmark network. The prediction confidence improves by 1.65-2.50% at three different dairy cow object scales, In addition, the keypoints under different occlusion conditions improve considerably, especially for small-scale keypoints, demonstrating the capability of the ASPP module for multi-scale feature extraction. By analyzing the distribution between the seven features and health, mild lameness, and severe lameness in dairy cows, it is shown that all the different features play an important role in distinguishing between different levels of lameness.

Genomic insights into ecological adaptation of oaks revealed by phylogenomic analysis of multiple species

Understanding the ecological adaptation of tree species can not only reveal the evolutionary potential but also benefit biodiversity conservation under global climate change. Quercus is a keystone genus in Northern Hemisphere forests, and its wide distribution in diverse ecosystems and long evolutionary history make it an ideal model for studying the genomic basis of ecological adaptations. Here we used a newly sequenced genome of Quercus gilva, an evergreen oak species from East Asia, with 18 published Fagales genomes to determine how Fagaceae genomes have evolved, identify genomic footprints of ecological adaptability in oaks in general, as well as between evergreen and deciduous oaks. We found that oak species exhibited a higher degree of genomic conservation and stability, as indicated by the absence of large-scale chromosomal structural variations or additional whole-genome duplication events. In addition, we identified expansion and tandem repetitions within gene families that contribute to plant physical and chemical defense (e.g., cuticle biosynthesis and oxidosqualene cyclase genes), which may represent the foundation for the ecological adaptation of oak species. Circadian rhythm and hormone-related genes may regulate the habits of evergreen and deciduous oaks. This study provides a comprehensive perspective on the ecological adaptations of tree species based on phylogenetic, genome evolutionary, and functional genomic analyses.

Multi-branch progressive embedding network for crowd counting

Crowd counting is essential for video surveillance and public safety. The performance of counting models has been greatly improved with the rapid development of Convolution neural networks (CNN), while it still suffers interference from complex background and large-scale variation. To relieve the above challenges, this paper proposes a novel Multi-branch Progressive Embedding Network (MPENet) for crowd counting. Specifically, the proposed network mainly consists of two modules named Background Area Filter (BAF) and Sequential Multi-scale Modules (SMM), which are embedded with each other to generate higher-quality density maps. Firstly, the BAF model base on attention mechanism is proposed to distinguish crowd from background, which effectively avoids the model outputting positive predictions in the background region. Meanwhile, a multi-level supervision mechanism is proposed to generate more accurate attention maps. Besides, the SMM module is designed to be progressively embedded with scale context information so that the scale feature will be smooth and continuous. Finally, a novel multi-scale consistency structural loss is proposed to avoid pixel-level isolation due to Euclidean loss. The proposed method significantly improves counting accuracy, achieving Mean Absolute Error (MAE) of 57.6 and 6.9 on ShanghaiTechA and ShanghaiTechB respectively.