Search Results (10,051)

In the realm of human–robot interaction, the integration of visual and verbal cues has become increasingly significant. This paper focuses on the challenges and advancements in referring image segmentation (RIS), a task that involves segmenting images based on textual descriptions. Traditional approaches to RIS have primarily focused on pixel-level classification. These methods, although effective, often overlook the interconnectedness of pixels, which can be crucial for interpreting complex visual scenes. Furthermore, while the PolyFormer model has shown impressive performance in RIS, its large number of parameters and high training data requirements pose significant challenges. These factors restrict its adaptability and optimization on standard consumer hardware, hindering further enhancements in subsequent research. Addressing these issues, our study introduces a novel two-branch decoder framework with SAM (segment anything model) for RIS. This framework incorporates an MLP decoder and a KAN decoder with a multi-scale feature fusion module, enhancing the model’s capacity to discern fine details within images. The framework’s robustness is further bolstered by an ensemble learning strategy that consolidates the insights from both the MLP and KAN decoder branches. More importantly, we collect the segmentation target edge coordinates and bounding box coordinates as input cues for the SAM model. This strategy leverages SAM’s zero-sample learning capabilities to refine and optimize the segmentation outcomes. Our experimental findings, based on the widely recognized RefCOCO, RefCOCO+, and RefCOCOg datasets, confirm the effectiveness of this method. The results not only achieve state-of-the-art (SOTA) performance in segmentation but are also supported by ablation studies that highlight the contributions of each component to the overall improvement in performance. Full article

In this paper, Panax ginseng cyclophilin (PgCyP) was successfully obtained through a genetic engineering technique. A bioinformatics method was used to analyze the physicochemical properties and structure of PgCyP. The results showed that PgCyP belongs to the cyclophilin gene family. The protein encoded by the PgCyP gene contains the active site of PPIase (R62, F67, and H133) and a binding site for cyclosporine A (W128). The relative molecular weight of PgCyP is 187.11 bp; its theoretical isoelectric point is 7.67, and it encodes 174 amino acids. The promoter region of PgCyP mainly contains the low-temperature environmental stress response (LTR) element, abscisic acid-responsive cis-acting element (ABRE), and light-responsive cis-acting element (G-Box). PgCyP includes a total of nine phosphorylation sites, comprising four serine phosphorylation sites, three threonine phosphorylation sites, and two tyrosine phosphorylation sites. PgCyP was recombined and expressed in vitro, and its recombinant expression was investigated. Furthermore, it was found that the recombinant PgCyP protein could effectively inhibit the germination of Phytophthora cactorum spores and the normal growth of Phytophthora cactorum mycelia in vitro. Further experiments on the roots of susceptible Arabidopsis thaliana showed that the PgCyP protein could improve the resistance of arabidopsis to Phytophthora cactorum. The findings of this study provide a basis for the use of the PgCyP protein as a new type of green biopesticide. Full article

Vehicle detection with optical remote sensing images has become widely applied in recent years. However, the following challenges have remained unsolved during remote sensing vehicle target detection. These challenges include the dense and arbitrary angles at which vehicles are distributed and which make it difficult to detect them; the extensive model parameter (Param) that blocks real-time detection; the large differences between larger vehicles in terms of their features, which lead to a reduced detection precision; and the way in which the distribution in vehicle datasets is unbalanced and thus not conducive to training. First, this paper constructs a small dataset of vehicles, MiVehicle. This dataset includes 3000 corresponding infrared and visible image pairs, offering a more balanced distribution. In the infrared part of the dataset, the proportions of different vehicle types are as follows: cars, 48%; buses, 19%; trucks, 15%; freight, cars 10%; and vans, 8%. Second, we choose the rotated box mechanism for detection with the model and we build a new vehicle detector, ML-Det, with a novel multi-scale feature fusion triple cross-criss FPN (TCFPN), which can effectively capture the vehicle features in three different positions with an mAP improvement of 1.97%. Moreover, we propose LKC–INVO, which allows involution to couple the structure of multiple large kernel convolutions, resulting in an mAP increase of 2.86%. We also introduce a novel C2F_ContextGuided module with global context perception, which enhances the perception ability of the model in the global scope and minimizes model Params. Eventually, we propose an assemble–disperse attention module to aggregate local features so as to improve the performance. Overall, ML-Det achieved a 3.22% improvement in accuracy while keeping Params almost unchanged. In the self-built small MiVehicle dataset, we achieved 70.44% on visible images and 79.12% on infrared images with 20.1 GFLOPS, 78.8 FPS, and 7.91 M. Additionally, we trained and tested our model on the following public datasets: UAS-AOD and DOTA. ML-Det was found to be ahead of many other advanced target detection algorithms. Full article

The Discrete Element Method (DEM) is an important tool for investigating the geotechnical properties of lunar regolith. The accuracy of DEM simulations largely depends on precise particle modeling and the appropriate selection of mesoscopic parameters. To enhance the reliability and accuracy of the DEM in lunar regolith studies, this paper utilized the high-fidelity IRSM-1 lunar regolith simulant to construct a DEM model with realistic particle shapes and conducted an angle of repose (AoR) simulation test. The optimal DEM parameters were calibrated using a combination of the Plackett–Burman test, steepest ascent test, and Box–Behnken design. The results indicate that the sliding friction coefficient, rolling friction coefficient, and surface energy significantly influence the simulation AoR. By optimizing against the measured AoR using a second-order regression model, the optimal parameter values were determined to be 0.633, 0.401, and 0.2, respectively. Under these optimal parameters, the error between the simulation and experimental AoR was 2.1%. Finally, the calibrated mesoscopic parameters were validated through a lifting cylinder test, showing an error of 6.3% between the simulation and experimental results. The high similarity in the shape of the AoR further confirms the accuracy and reliability of the parameter calibration method. This study provides a valuable reference for future DEM-based research on the mechanical and engineering properties of lunar regolith. Full article

The search for novel plant-based antioxidant and antibacterial medication has garnered a lot of attention lately. Glycyrrhiza glabra, known as licorice, is one of the most important medicinal plants. The primary component of Glycyrrhiza glabra is glycyrrhizin, which is biotransformed into 18α- and 18β-glycyrrhetinic acid for a variety of medicinal purposes. The goal of this study was to improve the bioavailability of glycyrrhizin by its biotransformation into glycyrrhetinic acid by Aspergillus niger. The biotransformation process was optimized using response surface methodology. A two-level Plackett–Burman design was employed to identify the factors that had a significant impact on the process of biotransformation. The three main variables were pH, glycerrhizin concentration, and incubation time. These three medium components were further optimized using a 3-level Box–Behnken design, and their optimum levels were pH of 8, an incubation period of 6 days, and a glycyrrhizin concentration of 1%. Using these optimum conditions, the maximum level obtained was 159% greater than in the screening experiment. Regarding the antimicrobial activity of glycyrrhizin extract, Bacillus subtilis emerged as the most sensitive organism with the lowest MIC (60 µg/mL) and the highest zone of inhibition (17 mm). The most resistant organism was Pseudomonas aeruginosa, which had the highest MIC (400 µg/mL) and the smallest zone of inhibition (10 mm). In the case of glycyrrhetinic acid, Bacillus subtilis was the most sensitive organism with the highest zone of inhibition (32 mm) and the lowest MIC (20 µg/mL). Pseudomonas aeruginosa was the most resistant organism, with the lowest zone of inhibition (18 mm), and the highest MIC (140 µg/mL). The antioxidant activity of glycyrrhizin extract increased from 12.81% at a concentration of 63 µg/100 µL to 41.41% at a concentration of 1000 µg/100 µL, while that of glycyrrhetinic acid extract increased from 35.5% at a concentration of 63 µg/100 µL to 76.85% at a concentration of 1000 µg/100 µL. The present study concluded that biotransformation of glycyrrhizin into glycyrrhetinic acid increased its bioavailability and antioxidant and antimicrobial activities. Glycyrrhizin and glycyrrhetinic acid might be used as a natural antimicrobial and antioxidant in pharmaceutical industries Full article

This study focused on the isolation of actinobacteria capable of producing extracellular keratinase from keratin-rich residues, which led to the selection of an actinobacterial strain referenced as Streptomyces strain DZ 06 (ES41). The Plackett–Burman screening plan was used for the statistical optimization of the enzymatic production medium, leading to the identification of five key parameters that achieved a maximum activity of 180.1 U/mL. Further refinement using response surface methodology (RSM) with a Box–Behnken design enhanced enzyme production to approximately 458 U/mL. Model validation, based on the statistical predictions, demonstrated that optimal keratinase activity of 489.24 U/mL could be attained with 6.13 g/L of chicken feather meal, a pH of 6.25, incubation at 40.65 °C for 4.11 days, and an inoculum size of 3.98 × 10⁷ spores/mL. The optimized culture conditions yielded a 21.67-fold increase in keratinase compared with the initial non-optimized standard conditions. The results show that this bacterium is an excellent candidate for industrial applications when optimal conditions are used to minimize the overall costs of the enzyme production process. Full article

Background: Busulfan is an FDA-approved alkylating drug used in the chemotherapy of advanced acute myeloid leukemia. The precise mechanisms by which Busulfan kills spermatogonia stem cells (SSCs) are not yet completely understood. Methods: Using a murine model, we evaluated Busulfan-induced apoptosis and DNA damage signaling between testis and ovary tissues. We executed RT-qPCR, analyzed single-nuclei RNA sequencing data and performed in situ hybridization for the localization of the gene expression in the tissues. Results: The results indicate that, in contrast to female germ cells, haploid male germ cells undergo significant apoptosis following Busulfan chemotherapy. Moreover, a gene enrichment analysis revealed that reactive oxygen species may activate the inflammatory response in part through the TNF-α/NF-κB signaling pathway. Interestingly, in the testis, the mRNA levels of TNF-α and TAF7 (TATA box-binding protein-associated factor 7) are downregulated, and testosterone levels suppressed. Mechanistically, the promoter of TNF-α has a conserved motif for binding TAF7, which is necessary for its transcriptional activation and may require further in-depth study. We next analyzed the tumorigenic function of TAF7 and revealed that it is highly overexpressed in several types of human cancers, particularly testicular germ cell tumors, and associated with poor patient survival. Therefore, we executed in situ hybridization and single-nuclei RNA sequencing, finding that less TAF7 mRNA is present in SSCs after chemotherapy. Conclusions: Thus, our data indicate a possible function of TAF7 in the regulation of SSCs and spermatogenesis following downregulation by Busulfan. These findings may account for the therapeutic effects of Busulfan and underlie its potential impact on cancer chemotherapy prognosis. Full article

Wind turbine blades (WTBs) are prone to damage from their working environment, including surface peeling and cracks. Early and effective detection of surface defects on WTBs can avoid complex and costly repairs and serious safety hazards. Traditional object detection methods have disadvantages of insufficient detection capabilities, extended model inference times, low recognition accuracy for small objects, and elongated strip defects within WTB datasets. In light of these challenges, a novel model named DCW-YOLO for surface damage detection of WTBs is proposed in this research, which leverages image data collected by unmanned aerial vehicles (UAVs) and the YOLOv8 algorithm for image analysis. Firstly, Dynamic Separable Convolution (DSConv) is introduced into the C2f module of YOLOv8, allowing the model to more effectively focus on the geometric structural details associated with damage on WTBs. Secondly, the upsampling method is replaced with the content-aware reassembly of features (CARAFE), which significantly minimizes the degradation of image characteristics throughout the upsampling process and boosts the network’s ability to extract features. Finally, the loss function is substituted with the WIoU (Wise-IoU) strategy. This strategy allows for a more accurate regression of the target bounding boxes and helps to improve the reliability in the localization of WTBs damages, especially for low-quality examples. This model demonstrates a notable superiority in surface damage detection of WTBs compared to the original YOLOv8n and has achieved a substantial improvement in the [email protected] metric, rising from 91.4% to 93.8%. Furthermore, in the more rigorous [email protected]–0.95 metric, it has also seen an increase from 68.9% to 71.2%. Full article

Creating an adult-sized humanoid robot with stable walking capabilities is a major challenge in robotics. While many renowned research groups focus on robots for perilous work environments and precision tasks, our approach simplifies balance control, making it accessible to robotics research groups and educational institutes. This facilitates the development of complex functionalities such as vision and object manipulation for adult-sized humanoids. This research article introduces RoboMan II, an advanced version of RoboMan I, which won the most prestigious award in all humanoid robot leagues at RoboCup 2016 due to its exceptional performance in walking and playing soccer. RoboMan II features significant improvements in performance, inherent stability, recovery after falls, and balance control. To facilitate its development, RoboMan II is lighter and incorporates a modified foot and parallel structure for its leg to boost its inherent stability, along with a two-stage balance control system for Immediate Response and Gradual Adaptation, enhancing its adaptability in various environments. Our simulation results demonstrate that RoboMan II’s walking stability on flat surfaces improved significantly in the face of minor perturbations, with the number of steps within the stable region increasing from 24%, with only the immediate controller to 58% when both controllers were used. Similar improvements were observed on inclined surfaces. Additionally, the 3D CAD files for all of the robot parts are released as open source in conjunction with this paper to facilitate reproduction and further innovation. The forthcoming RoboMan III will incorporate custom servo motors for increased speed, torque, and enhanced fall recovery, preventing disengagement of the gear box after a fall. It promises to be an invaluable asset for research and practical applications in humanoid robotics. Full article

Background: A pelletizer paired with hot-melt extrusion technology (HME) was used to develop colon-targeted pellets for ketoprofen (KTP). Thermal stability and side effects in the upper gastrointestinal tract made ketoprofen more suitable for this work. Methods: The pellets were prepared using the enzyme-triggered polymer Pectin LM in the presence of HPMC HME 4M, followed by pH-dependent Eudragit^® S 100 coating to accommodate the maximum drug release in the colon by minimizing drug release in the upper gastrointestinal tract (GIT). Box–Behnken Design (BBD) was used for response surface optimization of the proportion of different independent variables like Pectin LM (A), HPMC HME 4M (B), and Eudragit^® S 100 (C) required to lower the early drug release in upper GIT and to extend the drug release in the colon. Results: Solid-state characterization studies revealed that ketoprofen was present in a solid solution state in the hot-melt extruded polymer matrix. The desired responses of the prepared optimized KTP pellets obtained by considering the designed space showed 1.20% drug release in 2 h, 3.73% in the first 5 h of the lag period with the help of Eudragit^® S 100 coating, and 93.96% in extended release up to 24 h in the colonic region. Conclusions: Hence, developing Eudragit-coated hot-melt extruded pellets could be a significant method for achieving the colon-specific release of ketoprofen. Full article

TCP transcription factors are involved in the regulation of plant growth and development and response to stress. Previous studies showed that StTCP7 was involved in the abiotic stress response of potato and positively regulated plant tolerance to drought stress. On the basis of previous studies, this study verified the downstream target genes of StTCP7 transcription factor binding through yeast one hybridization, double luciferase and other technologies, and conducted a preliminary analysis of the downstream target genes. The results showed that the StTCP7 transcription factor could bind the promoter region of StDAM5 and StGOLS2 and regulate the expression of their genes. qRT-PCR analysis showed that the expression level of StDAM5 gene was the highest in flower stalk tissue and the lowest in leaf stalk. The expression of StGOLS2 gene was the highest in stem, the second in stalk, and the lower in root. Both StDAM5 and StGOLS2 genes responded to abiotic stress treated with 200 mM NaCl, 20% PEG-6000 and 100 µM ABA. The expression levels of target genes StDAM5 and StGOLS2 were up-regulated in StTCP7 interfered plants. The protein encoded by the target gene StDAM5 belongs to the Type II MADS-box protein, which contains 238 amino acids and is an acidic hydrophilic protein. The analysis of StDAM5 promoter region showed that the promoter region of StDAM5 gene contained cis-acting elements such as light response and abscisic acid. Subcellular localization showed that StDAM5 protein was expressed in both nucleus and cytoplasm. Full article

Bio-oil produced via fast pyrolysis, irrespective of the biomass source, faces several limitations, such as high water content, significant oxygenated compound concentration (35–40 wt.%), a low heating value (13–20 MJ/kg), and poor miscibility with fossil fuels. These inherent drawbacks hinder the bio-oil’s desirable properties and usability, highlighting the necessity for advanced processing techniques to overcome these challenges and improve the bio-oil’s overall quality and applicability in energy and industrial sectors. To address the limitations of bio-oil, a magnetic bimetallic oxide catalyst supported on activated rice straw biochar (ZrO₂-Fe₃O₄/AcB), which has not been previously employed for this purpose, was developed and characterized for upgrading rice straw bio-oil in supercritical butanol via esterification. Furthermore, the silica in the biochar, combined with the Lewis acid sites provided by ZrO₂ and Fe₃O₄, offers Brønsted acid sites. This synergistic combination enhances the bio-oil’s quality by facilitating esterification, deoxygenation, and mild hydrogenation, thereby reducing oxygen content and increasing carbon and hydrogen levels. The effects of variables, including time, temperature, and catalyst load, were optimized using response surface methodology (RSM). The optimal reaction conditions were determined using a three-factor, one-response, and three-level Box-Behnken design (BBD). The ANOVA results at a 95% confidence level indicate that the results are statistically significant due to a high Fisher’s test (F-value = 37.07) and a low probability (p-value = 0.001). The minimal difference between the predicted R² and adjusted R² for the ester yield (0.0092) suggests a better fit. The results confirm that the optimal reaction conditions are a catalyst concentration of 1.8 g, a reaction time of 2 h, and a reaction temperature of 300 °C. Additionally, the catalyst can be easily recycled for four reaction cycles. Moreover, the catalyst demonstrated remarkable reusability, maintaining its activity through four consecutive reaction cycles. Its magnetic properties allow for easy separation from the reaction mixture using an external magnet. Full article

This paper is concerned with an end-to-end electrocardiogram (ECG) signal transformation from a continuous-wave (CW) radar signal using a specialized deep learning model. For this purpose, the presented deep learning model is designed using convolutional neural networks (CNNs) and bidirectional long short-term memory (Bi-LSTM) with a maximum-overlap discrete wavelet transform (MODWT) layer and an adaptive neuro-fuzzy network (ANFN) layer. The proposed method has the advantage of developing existing deep networks and machine learning to reconstruct signals through CW radars to acquire ECG biological information in a non-contact manner. The fully connected (FC) layer of the CNN is replaced by an ANFN layer suitable for resolving black boxes and handling complex nonlinear data. The MODWT layer is activated via discrete wavelet transform frequency decomposition with maximum-overlap to extract ECG-related frequency components from radar signals to generate essential information. In order to evaluate the performance of the proposed model, we use a dataset of clinically recorded vital signs with a synchronized reference sensor signal measured simultaneously. As a result of the experiment, the performance is evaluated by the mean squared error (MSE) between the measured and reconstructed ECG signals. The experimental results reveal that the proposed model shows good performance in comparison to the existing deep learning model. From the performance comparison, we confirm that the ANFN layer preserves the nonlinearity of information received from the model by replacing the fully connected layer used in the conventional deep learning model. Full article

This study presents an innovative passive micromixer design featuring convergent–divergent sinusoidal walls, evaluated using the Villermaux–Dushman protocol. Five distinct designs were fabricated and tested, demonstrating superior mixing efficiency without additional obstructions. Testing of flow rates from 1000 to 50 mL/h revealed that the square-wave micromixer had the highest efficiency due to repeated fluid perturbations from its 90-degree angles. The loop-wave mixer performed the worst due to its lack of angles. The circular and box-wave mixers outperformed the loop-wave and backward arrow mixers due to their split and recombination effects. These designs, especially the circular and box-wave designs, offer optimal mixing for short-length applications, improving the efficiency and manufacturing simplicity for biomedical and biochemical analyses. Full article

The pursuit of the Sustainable Development Goals has highlighted rural electricity consumption patterns, necessitating innovative analytical approaches. This paper introduces a novel method for predicting rural electricity consumption by leveraging deep convolutional features extracted from satellite imagery. The study employs a pretrained remote sensing interpretation model for feature extraction, streamlining the training process and enhancing the prediction efficiency. A random forest model is then used for electricity consumption prediction, while the SHapley Additive exPlanations (SHAP) model assesses the feature importance. To explain the human geography implications of feature maps, this research develops a feature visualization method grounded in expert knowledge. By selecting feature maps with higher interpretability, the “black-box” model based on remote sensing images is further analyzed and reveals the geographical features that affect electricity consumption. The methodology is applied to villages in Xinxing County, Guangdong Province, China, achieving high prediction accuracy with a correlation coefficient of 0.797. The study reveals a significant positive correlations between the characteristics and spatial distribution of houses and roads in the rural built environment and electricity demand. Conversely, natural landscape elements, such as farmland and forests, exhibit significant negative correlations with electricity demand predictions. These findings offer new insights into rural electricity consumption patterns and provide theoretical support for electricity planning and decision making in line with the Sustainable Development Goals. Full article