Small object detection poses significant challenges in the realm of general object detection, primarily due to complex backgrounds and other instances interfering with the expression of features. This research introduces an uncomplicated and efficient algorithm that addresses the limitations of small object detection. Firstly, we propose an efficient cross-scale feature fusion attention module called ECFA, which effectively utilizes attention mechanisms to emphasize relevant features across adjacent scales and suppress irrelevant noise, tackling issues of feature redundancy and insufficient representation of small objects. Secondly, we design a highly efficient convolutional module named SEConv, which reduces computational redundancy while providing a multi-scale receptive field to improve feature learning. Additionally, we develop a novel dynamic focus sample weighting function called DFSLoss, which allows the model to focus on learning from both normal and challenging samples, effectively addressing the problem of imbalanced difficulty levels among samples. Moreover, we introduce Wise-IoU to address the impact of poor-quality examples on model convergence. We extensively conduct experiments on four publicly available datasets to showcase the exceptional performance of our method in comparison to state-of-the-art object detectors. Full article

Three-wheeled omnidirectional mobile robots (TOMRs) are widely used to accomplish precise transportation tasks in narrow environments owing to their stability, flexible operation, and heavy loads. However, these robots are susceptible to slippage. For wheeled robots, almost all faults and slippage will directly affect the power consumption. Thus, using the energy consumption model data and encoder data in the healthy condition as a reference to diagnose robot slippage and other system faults is the main issue considered in this paper. We constructed an energy model for the TOMR and analyzed the factors that affect the power consumption in detail, such as the position of the gravity center. The study primarily focuses on the characteristic relationship between power consumption and speed when the robot experiences slippage or common faults, including control system faults. Finally, we present the use of a table-based artificial neural network (ANN) to indicate the type of fault by comparing the modeled data with the measured data. The experiments proved that the method is accurate and effective for diagnosing faults in TOMRs. Full article

In this study, we explored the resilience of a clubroot resistance (CR) stacking model against a field population of Plasmodiophora brassicae pathotype 3H. This contrasts with our earlier work, where stacking CRaM and Crr1rutb proved only moderately resistant to pathotype X. Canola varieties carrying Rcr1/Crr1^rutb and Rcr1 + Crr1^rutb were repeatedly exposed to 3H at low (1 × 10⁴/g soil) and high (1 × 10⁷/g soil) initial resting spore concentrations over five planting cycles under controlled environments to mimic intensive canola production. Initially, all resistant varieties showed strong resistance. However, there was a gradual decline in resistance over time for varieties carrying only a single CR gene, particularly with Crr1^rutb alone and at the high inoculum level, where the disease severity index (DSI) increased from 9% to 39% over five planting cycles. This suggests the presence of virulent pathotypes at initially low levels in the 3H inoculum. In contrast, the variety with stacked CR genes remained resilient, with DSI staying below 3% throughout, even at the high inoculum level. Furthermore, the use of resistant varieties, carrying either a single or stacked CR genes, reduced the total resting spore numbers in soil over time, while the inoculum level either increased or remained high in soils where susceptible Westar was continuously grown. Our study demonstrates greater resistance resilience for stacking Rcr1 and Crr1^rutb against the field population of 3H. Additionally, the results suggest that resistance may persist even longer in fields with lower levels of inoculum, highlighting the value of extended crop rotation (reducing inoculum) alongside strategic CR-gene deployment to maximize resistance resilience. Full article

This paper proposes a new static output feedback control method to address the congestion control problem in transmission control protocol networks using active queue management routers. Based on linear matrix inequality optimization, this method determines a static output feedback control law to minimize the norm of the transfer function between the controlled queue length of the buffer and the exogenous disturbance affecting the available link bandwidth. A linear matrix inequality formulation is presented as a sufficient condition to guarantee the closed-loop system’s asymptotic stability while maintaining disturbance rejection within a specified level, regardless of round-trip time delays. The proposed robust static output feedback control eliminates the need to measure or estimate all system states, thus simplifying practical implementation. The effectiveness of the proposed design method is demonstrated by applying it in a practical process, as illustrated through a numerical example. Full article

Pingliang City has a complex topography and diverse soil types. To realize the improvement of soil according to local conditions and the reasonable and sustainable use of soil resources, an evaluation of soil quality in Pingliang City was carried out, based on the soil distribution situation in Pingliang City, adopting a method combining fuzzy mathematics and cluster analysis of the main evaluation factors, such as soil organic matter, topsoil depth, soil erosion intensity, soil moisture regime, effective soil thickness, soil texture, soil profile structure, soil nutrient status and topographical parts, to carry out a comprehensive evaluation. A comprehensive evaluation of soil quality was conducted in seven counties under the jurisdiction of Pingliang City, and the evaluation results were compared and analyzed against the national standard, “Cultivated land quality grade”, to provide a basis for the selection of scientific soil improvement methods. The results of the arable land quality grades indicate that the quality of farmland in Pingliang City is divided into three to ten grades, and the average quality grade of farmland is 6.83, which is in the middle–lower level, and the overall grade distribution shows the characteristics of low in the middle and high in the east and west. The results of fuzzy mathematics combined with cluster analysis indicated the following trends in soil quality for the 12 soil genera: Chuan black gunny soil > yellow moist soil > sandy soil > silt soil > mulching helilu soil> loessal soil> loamy soil > slope loessal soil > arenosol > tillage leaching gray cinnamon soil > calcareous gray cinnamon soil > red clay soil. The results of the combination of fuzzy mathematics and clustering were significantly correlated with the results of the evaluation of the soil quality of arable land; the correlation coefficient was 0.884. This indicates that the method can accurately and objectively review the advantages and disadvantages of arable land soil and can be effectively applied to the evaluation of the soil quality of agricultural soils in other regions. It is a complement to the existing evaluation of the soil quality of arable land and at the same time provides a reference for the improvement of soil quality in agricultural regions. Full article

This paper explores the emerging subject of extracting tannins from various plant sources using deep eutectic solvents (DESs). Tannins are widely used in the food and feed industries as they have outstanding antioxidant qualities and greatly enhance the flavor and nutritional content of a wide range of food products. Organic solvents are frequently used in traditional extraction techniques, which raises questions about their safety for human health and the environment. DESs present a prospective substitute because of their low toxicity, adaptability, and environmental friendliness. The fundamental ideas supporting the application of DESs in the extraction of tannins from a range of plant-based materials frequently used in daily life are all well covered in this paper. Furthermore, this paper covers the impact of extraction parameters on the yield of extracted tannins, as well as possible obstacles and directions for future research in this emerging subject. This includes challenges such as high viscosity, intricated recovery of compounds, thermal degradation, and the occurrence of esterification. An extensive summary of the diversity, structure, biosynthesis, distribution, and roles of tannins in plants is given in this paper. Additionally, this paper thoroughly examines various bioactivities of tannins and their metabolites. Full article

Bile acids help facilitate intestinal lipid absorption and have endocrine activity in glucose, lipid and bone metabolism. Obesity and exercise influence bile acid metabolism and have opposite effects in bone. This study investigates if regular exercise helps mitigate the adverse effects of obesity on bone, potentially by reversing alterations in bile acid metabolism. Four-month-old female Sprague Dawley rats either received a high-fat diet (HFD) or a chow-based standard diet (lean controls). During the 10-month study period, half of the animals performed 30 min of running at moderate speed on five consecutive days followed by two days of rest. The other half was kept inactive (inactive controls). At the study’s end, bone quality was assessed by microcomputed tomography and biomechanical testing. Bile acids were measured in serum and stool. HFD feeding was related to reduced trabecular (−33%, p = 1.14 × 10⁻⁷) and cortical (−21%, p = 2.9 × 10⁻⁸) bone mass and lowered femoral stiffness (12–41%, p = 0.005). Furthermore, the HFD decreased total bile acids in serum (−37%, p = 1.0 × 10⁻⁶) but increased bile acids in stool (+2-fold, p = 7.3 × 10⁻⁹). These quantitative effects were accompanied by changes in the relative abundance of individual bile acids. The concentration of serum bile acids correlated positively with all cortical bone parameters (r = 0.593–0.708), whilst stool levels showed inverse correlations at the cortical (r = −0.651–−0.805) and trabecular level (r = −0.656–−0.750). Exercise improved some trabecular and cortical bone quality parameters (+11–31%, p = 0.043 to 0.001) in lean controls but failed to revert the bone loss related to the HFD. Similarly, changes in bile acid metabolism were not mitigated by exercise. Prolonged HFD consumption induced quantitative and qualitative alterations in bile acid metabolism, accompanied by bone loss. Tight correlations between bile acids and structural indices of bone quality support further functional analyses on the potential role of bile acids in bone metabolism. Regular moderate exercise improved trabecular and cortical bone quality in lean controls but failed in mitigating the effects related to the HFD in bone and bile acid metabolism. Full article

Neurodegenerative diseases (NDs), characterized by progressive degeneration and death of neurons, are strongly related to aging, and the number of people with NDs will continue to rise. Alzheimer’s disease (AD) and Parkinson’s disease (PD) are the most common NDs, and the current treatments offer no cure. A growing body of research shows that AD and especially PD are intricately related to intestinal health and the gut microbiome and that both diseases can spread retrogradely from the gut to the brain. Zeolites are a large family of minerals built by [SiO₄]⁴⁻ and [AlO₄]⁵⁻ tetrahedrons joined by shared oxygen atoms and forming a three-dimensional microporous structure holding water molecules and ions. The most widespread and used zeolite is clinoptilolite, and additionally, mechanically activated clinoptilolites offer further improved beneficial effects. The current review describes and discusses the numerous positive effects of clinoptilolite and its forms on gut health and the gut microbiome, as well as their detoxifying, antioxidative, immunostimulatory, and anti-inflammatory effects, relevant to the treatment of NDs and especially AD and PD. The direct effects of clinoptilolite and its activated forms on AD pathology in vitro and in vivo are also reviewed, as well as the use of zeolites as biosensors and delivery systems related to PD. Full article

In many applications, such as atmospheric observation or disaster monitoring, cooperative control of a fleet of UAVs is crucial because it is effective in repeated tasks. In this work, we provide a workable and useful cooperative guiding algorithm for several fixed-wing UAVs to construct a path-following formation with communication delays. The two primary components of our concept are path-following (lateral guidance) and path formation (longitudinal guidance). The former is in charge of ensuring that, in the presence of wind disturbance, the lateral distance between the UAV and its targeted path converges using a well-known vector field technique. In the event of a communication delay, the latter ensures that several fixed-wing UAVs will create a predetermined formation shape. Furthermore, we provide a maximum delay bound that is dependent on the topology and a controller’s gain. Lastly, in order to confirm the viability and advantages of our suggested approach, we construct an effective platform for a hardware-in-the-loop (HIL) test. Full article

Deep learning has shown significant advantages in Automatic Dependent Surveillance-Broadcast (ADS-B) anomaly detection, but it is known for its susceptibility to adversarial examples which make anomaly detection models non-robust. In this study, we propose Time Neighborhood Accumulation Iteration Fast Gradient Sign Method (TNAI-FGSM) adversarial attacks which fully take into account the temporal correlation of an ADS-B time series, stabilize the update directions of adversarial samples, and escape from poor local optimum during the process of iterating. The experimental results show that TNAI-FGSM adversarial attacks can successfully attack ADS-B anomaly detection models and improve the transferability of ADS-B adversarial examples. Moreover, the TNAI-FGSM is superior to two well-known adversarial attacks called the Fast Gradient Sign Method (FGSM) and Basic Iterative Method (BIM). To the best of our understanding, we demonstrate, for the first time, the vulnerability of deep-learning-based ADS-B time series unsupervised anomaly detection models to adversarial examples, which is a crucial step in safety-critical and cost-critical Air Traffic Management (ATM). Full article

Given the prevalence of handwritten documents in human interactions, optical character recognition (OCR) for documents holds immense practical value. OCR is a field that empowers the translation of various document types and images into data that can be analyzed, edited, and searched. In handwritten recognition techniques, symmetry can be crucial to improving accuracy. It can be used as a preprocessing step to normalize the input data, making it easier for the recognition algorithm to identify and classify characters accurately. This review paper aims to summarize the research conducted on character recognition for handwritten documents and offer insights into future research directions. Within this review, the research articles focused on handwritten OCR were gathered, synthesized, and examined, along with closely related topics, published between 2019 and the first quarter of 2024. Well-established electronic databases and a predefined review protocol were utilized for article selection. The articles were identified through keyword, forward, and backward reference searches to comprehensively cover all relevant literature. Following a rigorous selection process, 116 articles were included in this systematic literature review. This review article presents cutting-edge achievements and techniques in OCR and underscores areas where further research is needed. Full article

Background and Objectives: This study aimed to elucidate the cytologic characteristics and diagnostic usefulness of endoscopic ultrasonography–fine needle aspiration cytology (EUS-FNAC) by comparing it with liquid-based preparation (LBP) and conventional smear (CS) in pancreas. Methods: The diagnostic categories (I through VII) were classified according to the World Health Organization Reporting System for Pancreaticobiliary Cytopathology. Ten cytologic features, including nuclear and additional features, were evaluated in 53 cases subjected to EUS-FNAC. Nuclear features comprised irregular nuclear contours, nuclear enlargement, hypochromatic nuclei with parachromatin clearing, and nucleoli. Additional cellular features included isolated atypical cells, mucinous cytoplasm, drunken honeycomb architecture, mitosis, necrotic background, and cellularity. A decision tree analysis was conducted to assess diagnostic efficacy. Results: The diagnostic concordance rate between LBP and CS was 49.1% (26 out of 53 cases). No significant differences in nuclear features were observed between categories III (atypical), VI (suspicious for malignancy), and VII (malignant). The decision tree analysis of LBP indicated that cases with moderate or high cellularity and mitosis could be considered diagnostic for those exhibiting nuclear atypia. Furthermore, in CS, mitosis, isolated atypical cells, and necrotic background exerted a more significant impact on the diagnosis of EUS-FNAC. Conclusions: Significant parameters for interpreting EUS-FNAC may differ between LBP and CS. While nuclear atypia did not influence the diagnosis of categories III, VI, and VII, other cytopathologic features, such as cellularity, mitosis, and necrotic background, may present challenges in diagnosing EUS-FNAC. Full article

There is a heightened risk of an oil spill occurring in the Arctic, as climate change driven sea ice loss permits an increase in Arctic marine transportation. The ability to detect an oil spill and monitor its progression is key to enacting an effective response. Microwave scatterometer systems may be used detect changes in sea ice thermodynamic and physical properties, so we examined the potential of C-band polarimetric radar for detecting diesel fuel beneath a thin sea ice layer. Sea ice physical properties, including thickness, temperature, and salinity, were measured before and after diesel addition beneath the ice. Time-series polarimetric C-band scatterometer measurements monitored the sea ice evolution and diesel migration to the sea ice surface. We characterized the temporal evolution of the diesel-contaminated seawater and sea ice by monitoring the normalized radar cross section (NRCS) and polarimetric parameters (conformity coefficient (μ), copolarization correlation coefficient (ρco)) at 20° and 25° incidence angles. We delineated three stages, with distinct NRCS and polarimetric results, which could be connected to the thermophysical state and the presence of diesel on the surface. Stage 1 described the initial formation of sea ice, while in Stage 2, we injected 20L of diesel beneath the sea ice. No immediate response was noted in the radar measurements. With the emergence of diesel on the sea ice surface, denoted by Stage 3, the NRCS dropped substantially. The largest response was for VV and HH polarizations at 20° incidence angle. Physical sampling indicated that diesel emerged to the surface of the sea ice and trended towards the tub edge and the polarimetric scatterometer was sensitive to these physical changes. This study contributes to a greater understanding of how C-band frequencies can be used to monitor oil products in the Arctic and act as a baseline for the interpretation of satellite data. Additionally, these findings will assist in the development of standards for oil and diesel fuel detection in the Canadian Arctic in association with the Canadian Standards Association Group. Full article

Blue light is an important light wavelength in regulating plant flowering. In a controlled environment (CE) plant production systems, blue light can be manipulated easily and even precisely through electric lighting, especially with the advancement of light-emitted diode (LED) technologies. However, the results of previous studies in the literature about blue-light-mediated flowering are inconsistent, which would limit its practical application in CE plant production while implying that an in-depth study of the relevant physiological mechanism is necessary in the future. This review consolidates and analyzes the diverse findings from previous studies on blue light-mediated plant flowering in varying high-value crops from ornamental plants to fruits, vegetables, and specialty crops. By synthesizing the contrasting results, we proposed the possible explanations and even the underlying mechanisms related to blue light intensity and exposure duration, its co-action with other light wavelengths, background environment conditions, and the involved photoreceptors. We have also identified the knowledge gaps based on these studies and outlined future directions for research and potential application in this promising field. This review provides valuable insights into the important and diverse role of blue light in plant flowering and offers a foundation for further investigations to optimize plant flowering through lighting technologies. Full article

Background: Materialism is an attitude that considers material goods to be central in life. Nowadays, adolescents appear to have a high level of materialism, which is related to risky behaviors. Nevertheless, there is a lack of measurement tools with adequate psychometric properties to assess materialism in this age group. For this reason, two studies were conducted to investigate the psychometric properties of the original and short Material Values Scale (MVS) in adolescents. Methods: In Study 1, participants were randomly split into two subsamples to compare psychometric properties of the original version of MVS with those of the short one. The first subsample consisted of 1054 adolescents (58% male; Mage = 16.34; SD = 1.15), and the second one of 1058 adolescents (57% male; Mage = 16.26; SD = 1.04). In Study 2, the psychometric properties of a revised version of the short MVS (without item 8) were investigated to confirm its adequacy with a new sample composed of 1896 adolescents (60% male; Mage = 16.40; SD = 2.76). Results: Results of Study 1 showed that the short version appeared to be a better measuring tool with respect to the long form to investigate materialism in adolescents. Nevertheless, problems with item 8 emerged. Results of Study 2 attested to the adequacy of the psychometric properties of the revised version of the short MVS (by excluding item 8) in this age group, in terms of dimensionality, reliability, and validity. Conclusions: Findings show that the revised short version of the MVS could be a valid and reliable tool for measuring the multidimensional construct of materialism in Italian adolescents. Full article

This paper presents research results on biocomposites made from a combination of extruded apple pomace (EAP) and potato starch (SP). The aim of this work was to investigate the basic properties of biocomposites obtained from extruded apple pomace reinforced with potato starch. The products were manufactured by hot pressing using a hydraulic press with a mould for producing samples. The prepared biocomposites were subjected to strength tests, surface wettability was determined, and a colour analysis was carried out. A thermogravimetric analysis (TGA), Fourier transform infrared spectroscopy (FTIR), and cross-sectioning observed in a scanning electron microscope (SEM) were also performed. The obtained test results showed that the combination of apple pomace (EAP) and starch (SP) enabled the production of compact biocomposite materials. At the same time, it was found that each increase in the share of starch in the mixture for producing biocomposites increased the strength parameters of the obtained materials. With the highest share of starch in the mixture, 40%, and a raw material moisture content of 14%, the material had the best strength parameters and was even characterised by hydrophobic properties. It was also found that materials with a high content of starch are characterised by increased temperature resistance. The analysis of SEM microscopic photos showed well-glued particles of apple pomace, pectin, and gelatinised starch and a smooth external structure of the samples. Research and analyses have shown that apple pomace reinforced only with the addition of starch can be a promising raw material for the production of simple, biodegradable biocomposite materials. Full article

The importance of forecasting crop yields in agriculture cannot be overstated. The effects of yield forecasting are observed in all the aspects of the supply chain from staffing to supplier demand, food waste, and other business decisions. However, the process is often inaccurate and far from perfect. This paper explores the potential of using expert forecasts to enhance the crop yield predictions of our global-to-local XGBoost machine learning system. Additionally, it investigates the ERA5 climate model’s viability as an alternative data source for crop yield forecasting in the absence of on-farm weather data. We find that, by combining both the expert’s pre-season forecasts and the ERA5 climate model with the machine learning model, we can—in most cases—obtain better forecasts that outperform the growers’ pre-season forecasts and the machine learning-only models. Our expert-informed model attains yield forecasts for 4 weeks ahead with an average RMSE of 0.0855 across all the plots and an RMSE of 0.0872 with the ERA5 climate data included. Full article

Three-phase gliding arc discharge reactors are devices in which it is difficult to maintain stable plasma parameters, be it electrically, physically, or chemically. The main cause of plasma instability is the source, which is freely burning arcs in a three-phase system. In addition, these arcs burn at low currents and are intensively cooled, further increasing their instability. These instabilities translate into the electrical characteristics of the plasma reactor. The analysis for the four gases nitrogen, argon, helium, and air shows that the type of plasma-generating gas and its physical parameters have a strong influence on the operational characteristics of the plasma reactor. Current–voltage, power and frequency characteristics of the plasma reactor were plotted experimentally. Characteristics obtained in this way make it possible to determine the areas of effective operation of the plasma reactor, and to estimate the quality of the generated plasma. Based on the characteristics obtained, a method of controlling the plasma parameters can be developed. Full article

The Grand Canal possesses a unique ecosystem as one of the world cultural heritage sites. However, its ecological roles and services have been underemphasized in heritage conservation efforts, leading to environmental pollution and the degradation of its heritage value, especially in the highly urbanized southern Jiangsu section downstream. This study examines the synergy between regulating ecosystem services (RESs) and cultural ecosystem services (CESs) along the highly urbanized Wuxi section of the Grand Canal, as well as the environmental drivers influencing this relationship. The findings reveal that the synergy between CESs and RESs does not occur at the urban–rural interface but rather in areas 4-6 km away from the Central Business District (CBD). Certain drivers, such as floor area ratio (FAR) and plot ratio (PR), enhance both RES and CES within specific sections. Beyond riparian green belts, various low-density, dispersed, and well-vegetated lands could serve as significant contributors to future ecosystem service (ES) synergy. These conclusions further elucidate the spatial distribution of the heritage cultural services and other ES synergy along the Grand Canal, providing scientific support for broader improvements guiding the sustainable co-development in similar urbanized areas. Full article

Increasing global credit card usage has elevated it to a preferred payment method for daily transactions, underscoring its significance in global financial cybersecurity. This paper introduces a credit card fraud detection (CCFD) system that integrates federated learning (FL) with blockchain technology. The experiment employs FL to establish a global learning model on the cloud server, which transmits initial parameters to individual local learning models on fog nodes. With three banks (fog nodes) involved, each bank trains its learning model locally, ensuring data privacy, and subsequently sends back updated parameters to the global learning model. Through the integration of FL and blockchain, our system ensures privacy preservation and data protection. We utilize three machine learning and deep neural network learning algorithms, RF, CNN, and LSTM, alongside deep optimization techniques such as ADAM, SGD, and MSGD. The SMOTE oversampling technique is also employed to balance the dataset before model training. Our proposed framework has demonstrated efficiency and effectiveness in enhancing classification performance and prediction accuracy. Full article

The geogrid flexible reinforced soil wall is widely used in engineering practice. However, a more comprehensive understanding of the dynamic behavior of reinforced soil wall is still required for a more reasonable application. In order to explore the mechanical behavior of a geogrid flexible reinforced soil wall, the model test was carried out to investigate the dynamic deformation of geogrid reinforced soil wall subjected to a repeated load. The numerical simulation was also conducted for comparison and extension with regards to the earth pressure and the reinforcement strain. The change rules for the deformation of the wall face, the vertical earth pressure and the reinforcement strain subjected to dynamic load with four frequencies (4, 6, 8 and 10 Hz) and four amplitudes (30–60, 40–80, 50–100 and 60–120 kPa) were obtained. The factors that affect the mechanical behavior of geogrid flexible reinforced soil wall were analyzed. The results show that the dynamic deformation characteristics of reinforced soil wall are affected by the number of vibrations, the amplitude of dynamic load and the frequency of vibration. The maximum lateral displacement of the reinforced soil wall occurs on the third to the fifth layer. With an increase in dynamic load amplitude, the development of dynamic deformation gradually increases, and after a cumulative vibration of 200 × 10⁴ times, the cumulative lateral deformation ratio and the cumulative vertical deformation ratio of the wall face is less than 1%. The vertical earth pressure of geogrid flexible reinforced soil wall increases partially along the length of the reinforcement, and the vertical earth pressure of the third layer is basically unchanged when subjected to a dynamic load. With an increase in vibration number, the change in the reinforcement strain of the third layer is more complex, and the change rules of the reinforcement strain of each layer are different. The reinforcement strain is small, with a maximum value of 0.1%. Full article

Graph neural networks (GNNs) are crucial tools for processing non-Euclidean data. However, due to scalability issues caused by the dependency and topology of graph data, deploying GNNs in practical applications is challenging. Some methods aim to address this issue by transferring GNN knowledge to MLPs through knowledge distillation. However, distilled MLPs cannot directly capture graph structure information and rely only on node features, resulting in poor performance and sensitivity to noise. To solve this problem, we propose a lightweight optimization method for GNNs that combines graph contrastive learning and variable-temperature knowledge distillation. First, we use graph contrastive learning to capture graph structural representations, enriching the input information for the MLP. Then, we transfer GNN knowledge to the MLP using variable temperature knowledge distillation. Additionally, we enhance both node content and structural features before inputting them into the MLP, thus improving its performance and stability. Extensive experiments on seven datasets show that the proposed KDGCL model outperforms baseline models in both transductive and inductive settings; in particular, the KDGCL model achieves an average improvement of 1.63% in transductive settings and 0.8% in inductive settings when compared to baseline models. Furthermore, KDGCL maintains parameter efficiency and inference speed, making it competitive in terms of performance. Full article