Search Results (3,429)

Social networks have become deeply integrated into our daily lives, leading to an increase in image sharing across different platforms. Simultaneously, the existence of robust and user-friendly media editors not only facilitates artistic innovation, but also raises concerns regarding the ease of creating misleading media. This highlights the need for developing new advanced techniques for the image copy detection task, which involves evaluating whether photos or videos originate from the same source. This research introduces a novel application of the Vision Transformer (ViT) model to the image copy detection task on the DISC21 dataset. Our approach involves innovative strategic sampling of the extensive DISC21 training set using K-means clustering to achieve a representative subset. Additionally, we employ complex augmentation pipelines applied while training with varying intensities. Our methodology follows the instance discrimination concept, where the Vision Transformer model is used as a classifier to map different augmentations of the same image to the same class. Next, the trained ViT model extracts descriptors of original and manipulated images that subsequently underwent post-processing to reduce dimensionality. Our best-achieving model, tested on a refined query set of 10K augmented images from the DISC21 dataset, attained a state-of-the-art micro-average precision of 0.79, demonstrating the effectiveness and innovation of our approach. Full article

At present, stationary phases based on ionic liquids are a promising and widely used technique in gas chromatography, yet they remain poorly studied. Unfortunately, testing of “new” stationary phases is often carried out on a limited set of test compounds (about 10 compounds) of relatively simple structures. This study represents the first investigation into the physicochemical patterns of retention of substituted (including polysubstituted) aromatic alcohols on two stationary phases of different polarities: one based on pyridinium-based ionic liquids and the other on a standard polar phase. The retention order of the studied compounds on such stationary phases compared to the standard polar phase, polyethylene glycol (SH-Stabilwax), was compared and studied. It was shown that pyridinium-based ionic liquids stationary phase has a different selectivity compared to the SH-Stabilwax. Using a quantitative structure–retention relationships (QSRR) study, the differences in selectivity of the two stationary phases were interpreted. Using CHERESHNYA software, the importance of descriptors on different stationary phases was evaluated for the same data set. Different selectivity of the stationary phases correlates with different contributions of descriptors for the analytes under study. For the first time, we show that in-column dehydration is observed for some compounds (mostly substituted benzyl alcohols). This effect is worthy of further investigation and requires attention when analyzing complex mixtures. It suggests that when testing “new” stationary phases, it is necessary to conduct tests on a large set of different classes of compounds. This is because, in the case of using ionic liquids as an stationary phase, a reaction between the analyte and the stationary phase is possible. Full article

In this paper, a stability analysis and the controller improvement of T-S fuzzy Descriptor system are studied. Firstly, by making full use of the related theory of fuzzy affiliation function and combining the design method of fuzzy Lyapunov function with the method of inequality deflation, a stability condition with wider admissibility and less system conservatism is proposed. The advantage of this method is that it is not necessary to ensure that each fuzzy subsystem is progressively stable. We also maximise the boundary of the derivatives of the affiliation function mined. Secondly, a PDC controller and a Non-PDC controller are designed, and the deflation conditions for the linear matrix inequalities of the two controllers are constructed. Finally, some arithmetic simulations and practical examples are given to demonstrate the effectiveness of the method studied in this paper, and the results obtained are less conservative and have larger feasible domains than previous methods. Full article

The atomistic structure and point-defect thermodynamics of the model Σ5(310)[001] grain boundary in CeO₂ were explored with atomistic simulations. An interface with a double-diamond-shaped structural repeat unit was found to have the lowest energy. Segregation energies were calculated for oxygen vacancies, electron polarons, gadolinium and scandium acceptor cations, and tantalum donor cations. These energies deviate strongly from their bulk values over the same length scale, thus indicating a structural grain-boundary width of approximately 1.5 nm. However, an analysis revealed no unambiguous correlation between segregation energies and local structural descriptors, such as interatomic distance or coordination number. From the segregation energies, the grain-boundary space-charge potential in Gouy–Chapman and restricted-equilibrium regimes was calculated as a function of temperature for dilute solutions of (i) oxygen vacancies and acceptor cations and (ii) electron polarons and donor cations. For the latter, the space-charge potential is predicted to change from negative to positive in the restricted-equilibrium regime. For the former, the calculation of the space-charge potential from atomistic segregation energies is shown to require the inclusion of the segregation energies for acceptor cations. Nevertheless, the space-charge potential in the restricted-equilibrium regime can be described well with an empirical model employing a single effective oxygen-vacancy segregation energy. Full article

Semantic segmentation of target objects in power transmission line corridor point cloud scenes is a crucial step in powerline tree barrier detection. The massive quantity, disordered distribution, and non-uniformity of point clouds in power transmission line corridor scenes pose significant challenges for feature extraction. Previous studies have often overlooked the core utilization of spatial information, limiting the network’s ability to understand complex geometric shapes. To overcome this limitation, this paper focuses on enhancing the deep expression of spatial geometric information in segmentation networks and proposes a method called BDF-Net to improve RandLA-Net. For each input 3D point cloud data, BDF-Net first encodes the relative coordinates and relative distance information into spatial geometric feature representations through the Spatial Information Encoding block to capture the local spatial structure of the point cloud data. Subsequently, the Bilinear Pooling block effectively combines the feature information of the point cloud with the spatial geometric representation by leveraging its bilinear interaction capability thus learning more discriminative local feature descriptors. The Global Feature Extraction block captures the global structure information in the point cloud data by using the ratio between the point position and the relative position, so as to enhance the semantic understanding ability of the network. In order to verify the performance of BDF-Net, this paper constructs a dataset, PPCD, for the point cloud scenario of transmission line corridors and conducts detailed experiments on it. The experimental results show that BDF-Net achieves significant performance improvements in various evaluation metrics, specifically achieving an OA of 97.16%, a mIoU of 77.48%, and a mAcc of 87.6%, which are 3.03%, 16.23%, and 18.44% higher than RandLA-Net, respectively. Moreover, comparisons with other state-of-the-art methods also verify the superiority of BDF-Net in point cloud semantic segmentation tasks. Full article

The phenological and morphological variation studies among forms of Laggera Sch. Bip. Ex. Benth. and Hook species are limited, despite the medicinal use of the genus. Therefore, this study aimed to document phenology and morphological variation in cultivated populations of Laggera alata and Laggera crispata forms from seedling to maturity. The forms were categorized as Laggera alata with a small capitulum (LA-SC), Laggera alata with a large capitulum (LA-BC), Laggera crispata from South Africa (LC-SA), and Laggera crispata from Zimbabwe (LC-ZIM). Seeds were germinated in Petri dishes, transplanted to plug trays, and later to field plots at 60 days. Phenological events were recorded when observed in at least one plant. Twelve qualitative and four morphometric traits were measured monthly on five plants per Laggera form. Analysis of variance and Tukey’s Honestly Significant Difference test (p < 0.05) were used for data analysis. Results indicated significant variation in phenology, qualitative traits, leaf traits, plant height, and stem diameter both within and between L. crispata and L. alata forms. Morphometric traits, such as leaf size and the number of leaves per plant, were identified as key descriptors for differentiating L. alata forms. These findings provide a foundation for the introduction of Laggera forms into farming systems for medicinal and commercial purposes. Full article

Tuberculosis (TB) is an ancient global public health problem. Several strategies have been applied to develop new and more effective vaccines against TB, from attenuated or inactivated mycobacteria to recombinant subunit or genetic vaccines, including viral vectors. This review aimed to evaluate patents filed between 2010 and 2023 for TB vaccine candidates. It focuses on viral vector-based strategies. A search was carried out in Espacenet, using the descriptors “mycobacterium and tuberculosis” and the classification A61K39. Of the 411 patents preliminarily identified, the majority were related to subunit vaccines, with 10 patents based on viral vector platforms selected in this study. Most of the identified patents belong to the United States or China, with a concentration of patent filings between 2013 and 2023. Adenoviruses were the most explored viral vectors, and the most common immunodominant Mycobacterium tuberculosis (Mtb) antigens were present in all the selected patents. The majority of patents were tested in mouse models by intranasal or subcutaneous route of immunization. In the coming years, an increased use of this platform for prophylactic and/or therapeutic approaches for TB and other diseases is expected. Along with this, expanding knowledge about the safety of this technology is essential to advance its use. Full article

This study explores the application of fine-tuned large language models for predicting physicochemical properties, specifically focusing on Abraham model solute descriptors (E, S, A, B, V) and modified solvent parameters (e₀, s₀, a₀, b₀, v₀). By leveraging ChemLLaMA, a specialized version of the LLaMA model for cheminformatics tasks, we developed the AbraLlama-Solvent and AbraLlama-Solute models using curated datasets of experimentally derived solute descriptors and solvent parameters. Our findings demonstrate that AbraLlama-Solvent and AbraLlama-Solute predict modified solvent parameters and solute descriptors with high accuracy, comparable to existing methods. The AbraLlama-Solvent model shows varying prediction accuracy across different solvents, influenced by their position within the chemical space, while the AbraLlama-Solute model consistently predicts solute descriptors with high accuracy. Both models are available as applications on Hugging Face, facilitating easy predictions from SMILES strings. This research highlights the potential of LLMs in chemistry applications, offering practical tools for solvent comparison and expanding the applicability of Abraham solvation equations to a broader range of organic solvents. Full article

This paper investigates the relationship between syllabic duration and F0 contours for implementing three prosodic functions. Work on rhythm usually describes the evolution of syllable-sized durations throughout utterances, rarely making reference to melodic events. On the other hand, work on intonation usually describes linear sequences of melodic events with indirect references to duration. Although some scholars have explored the relationship between these two parameters for particular functions, to our knowledge, there has been no investigation on the systematic correlation between syllabic duration and F0 values throughout narrative sequences. Based on a corpus of story retelling with nine speakers of Brazilian Portuguese from two regions, our work investigated the interplay between syllabic duration and melody to signal three prosodic functions: terminal and non-terminal boundary marking and prominence. The examination of local syllabic duration maxima and four F0 descriptors revealed that these maxima act as landmarks for particular F0 shapes: for non-terminal boundaries, the great majority of shapes were increasing and increasing–decreasing patterns; for terminal boundaries, almost all shapes were decreasing F0 patterns; and for prominence marking, the great majority of shapes were high tones across the stressed syllable. Time series analyses revealed significant correlations between duration and specific F0 descriptors, pointing to a ruled interplay between F0 and syllabic duration patterns in Brazilian Portuguese story retelling. Full article

The degradation of concrete and reinforced concrete structures is a significant technical and economic challenge, requiring continuous repair and rehabilitation throughout their service life. Geopolymers (GPs), known for their high mechanical strength, low shrinkage, and durability, are being increasingly considered as alternatives to traditional repair materials. However, there is currently a lack of understanding regarding the interface bond properties between new geopolymer layers and old concrete substrates. In this paper, using advanced computational techniques, including quantum mechanical calculations and stochastic modeling, we explored the adsorption behavior and interaction mechanism of aluminosilicate oligomers with different Si/Al ratios forming the geopolymer gel structure and calcium silicate hydrate as the substrate at the interface bond region. We analyzed the electron density distributions of the highest occupied and lowest unoccupied molecular orbitals, examined the reactivity indices based on electron density functional theory, performed Mulliken charge population analysis, and evaluated global reactivity descriptors for the considered oligomers. The results elucidate the mechanisms of local and global reactivity of the oligomers, the equilibrium low-energy configurations of the oligomer structures adsorbed on the surface of C-(A)-S-H(I) (100), and their adsorption energies. These findings contribute to a better understanding of the adhesion properties of geopolymers and their potential as effective repair materials. Full article

In case of future viral threats, including the proposed Disease X that has been discussed since the emergence of the COVID-19 pandemic in March 2020, our research has focused on the development of antiviral strategies using fragrance compounds with known antiviral activity. Despite the recognized antiviral properties of mixtures of certain fragrance compounds, there has been a lack of a systematic approach to optimize these mixtures. Confronted with the significant combinatorial challenge and the complexity of the compound formulation space, we employed Bayesian optimization, guided by Gaussian Process Regression (GPR), to systematically explore and identify formulations with demonstrable antiviral efficacy. This approach required the transformation of the characteristics of formulations into quantifiable feature values using molecular descriptors, subsequently modeling these data to predict and propose formulations with likely antiviral efficacy enhancements. The predicted formulations underwent experimental testing, resulting in the identification of combinations capable of inactivating 99.99% of viruses, including a notably efficacious formulation of five distinct fragrance types. This model demonstrates high predictive accuracy (coefficient determination Rcv2 > 0.7) and suggests a new frontier in antiviral strategy development. Our findings indicate the powerful potential of computational modeling to surpass human analytical capabilities in the pursuit of complex, fragrance-based antiviral formulations. Full article

Photoresponsive drug delivery stands as a pivotal frontier in smart drug administration, leveraging the non-invasive, stable, and finely tunable nature of light-triggered methodologies. The generative pre-trained transformer (GPT) has been employed to generate molecular structures. In our study, we harnessed GPT-2 on the QM7b dataset to refine a UV-GPT model with adapters, enabling the generation of molecules responsive to UV light excitation. Utilizing the Coulomb matrix as a molecular descriptor, we predicted the excitation wavelengths of these molecules. Furthermore, we validated the excited state properties through quantum chemical simulations. Based on the results of these calculations, we summarized some tips for chemical structures and integrated them into the alignment of large-scale language models within the reinforcement learning from human feedback (RLHF) framework. The synergy of these findings underscores the successful application of GPT technology in this critical domain. Full article

Point cloud registration is a crucial technique in photogrammetry, remote sensing, etc. A generalized 3D point cloud registration framework has been developed to estimate the optimal rigid transformation between two point clouds using 3D key point correspondences. However, challenges arise due to the uncertainty in 3D key point detection techniques and the similarity of local surface features. These factors often lead to feature descriptors establishing correspondences containing significant outliers. Current point cloud registration algorithms are typically hindered by these outliers, affecting both their efficiency and accuracy. In this paper, we propose a fast and robust point cloud registration method based on a compatibility graph and accelerated guided sampling. By constructing a compatible graph with correspondences, a minimum subset sampling method combining compatible edge sampling and compatible vertex sampling is proposed to reduce the influence of outliers on the estimation of the registration parameters. Additionally, an accelerated guided sampling strategy based on preference scores is presented, which effectively utilizes model parameters generated during the iterative process to guide the sampling toward inliers, thereby enhancing computational efficiency and the probability of estimating optimal parameters. Experiments are carried out on both synthetic and real-world data. The experimental results demonstrate that our proposed algorithm achieves a significant balance between registration accuracy and efficiency compared to state-of-the-art registration algorithms such as RANSIC and GROR. Even with up to 2000 initial correspondences and an outlier ratio of 99%, our algorithm achieves a minimum rotation error of 0.737° and a minimum translation error of 0.0201 m, completing the registration process within 1 s. Full article

In this paper, an alignment-free bioinformatics technique, termed the 20D-Dynamic Representation of Protein Sequences, is utilized to investigate the similarity/dissimilarity between Baculovirus and Echinococcus multilocularis genome sequences. In this method, amino acid sequences are depicted as 20D-dynamic graphs, comprising sets of “material points” in a 20-dimensional space. The spatial distribution of these material points is indicative of the sequence characteristics and is quantitatively described by sequence descriptors akin to those employed in dynamics, such as coordinates of the center of mass of the 20D-dynamic graph and the tensor of the moment of inertia of the graph (defined as a symmetric matrix). Each descriptor unveils distinct features of similarity and is employed to establish similarity relations among the examined sequences, manifested either as a symmetric distance matrix (“similarity matrix”), a classification map, or a phylogenetic tree. The classification maps are introduced as a new way of visualizing the similarity relations obtained using the 20D-Dynamic Representation of Protein Sequences. Some classification maps are obtained using the Principal Component Analysis (PCA) for the center of mass coordinates and normalized moments of inertia of 20D-dynamic graphs as input data. Although the method operates in a multidimensional space, we also apply some visualization techniques, including the projection of 20D-dynamic graphs onto a 2D plane. Studies on model sequences indicate that the method is of high quality, both graphically and numerically. Despite the high similarity observed among the sequences of E. multilocularis, subtle discrepancies can be discerned on the 2D graphs. Employing this approach has led to the discovery of numerous new similarity relations compared to our prior study conducted at the DNA level, using the 4D-Dynamic Representation of DNA/RNA Sequences, another alignment-free bioinformatics method also introduced by us. Full article

Phaseolus species are cultivated worldwide as a primary food source for human consumption. Common bean (Phaseolus vulgaris L.) and runner bean (Phaseolus coccineus L.) landraces are often cultivated together. The purpose was to document the traditional knowledge held by farmers regarding bean landraces, to describe the diversity through seed morphological descriptors (five quantitative and five qualitative traits), to understand the geographic distribution using Moran’s I statistic, and to analyze the diversity through the Shannon–Wiener Diversity–Equity (H′) index and the Shannon J’ homogeneity index at the community level to better understand the richness of this geographic area. A total of 361 common and runner bean accessions were collected in 10 Andean rural communities of Cotacachi, North Ecuador. We identified 47 landraces, predominantly kept by indigenous female farmers, with limited access to formal education, highlighting the role of this disadvantaged population in agrobiodiversity conservation. The cluster analysis revealed three groups (cophenetic correlation coefficient = 0.6). In the principal component (PC) analysis, 94% of the variation was explained by PC1 and PC2. A positive spatial autocorrelation (Moran’s I: 0.24; z-score: 2.20; p-value: 0.03) was identified, confirming a spatial structure. The Morochos community showed the highest diversity (H′ = 1.55). The information of the diversity and distribution of common and runner bean variability provided in this study is a contribution to further research focused on conservation. Full article