Valuable Applications Research Articles

Unambiguous discrimination of general quantum operations

The discrimination of quantum operations has long been an intriguing challenge, with theoretical research notably advancing our understanding of the quantum features in discriminating quantum objects. This challenge is closely related to the discrimination of quantum states, and proof-of-principle demonstrations of the latter have already been realized using optical photons. However, the experimental demonstration of discriminating general quantum operations, including both unitary and nonunitary operations, has remained elusive. In general quantum systems, especially those with high dimensions, the preparation of arbitrary quantum states and the implementation of arbitrary quantum operations and generalized measurements are nontrivial tasks. Here, we experimentally demonstrate the optimal unambiguous discrimination of up to six displacement operators and the unambiguous discrimination of nonunitary quantum operations. Our results demonstrate powerful tools for experimental research in quantum information processing and are expected to stimulate a wide range of valuable applications in the field of quantum sensing.

Utilization of Alternative Fibres Manufactured from Recycled PET Bottles in Concrete Technology for the Improvement of Fire Resistance

This article addresses the potential use of secondary polymer fibres in the field of structural concrete as a replacement for primary polymer fibres (mainly polypropylene/PP/), which are used in concrete to enhance its resistance when exposed to high temperatures (especially in the case of fire). Research has shown that, in addition to PP fibres, polyethylene terephthalate/PET/fibres, produced by recycling packaging materials (mainly PET bottles), can also be used as an alternative. These fibres are industrially produced in similar dimensions as PP fibres and exhibit similar behaviour when added to fresh and hardened concrete. In terms of their effect on increasing resistance to extreme heat loads, it has been found that despite a higher melting point (Tm), concrete with these fibres demonstrates comparable fire resistance. Therefore, it can be concluded that secondary PET fibres represent an interesting alternative to primary PP fibres from the perspective of a circular economy, and their use in construction represents a potentially valuable application for PET obtained through the collection and recycling of PET packaging materials.

An Ultra‐Short‐Term Multi‐Step Prediction Model for Wind Power Based on Sparrow Search Algorithm, Variational Mode Decomposition, Gated Recurrent Unit, and Support Vector Regression

ABSTRACTAccurate ultra‐short‐term wind power prediction techniques are crucial for ensuring the efficient and safe operation of wind farms and power systems. Combined models based on data decomposition‐prediction techniques have shown excellent performance in ultra‐short‐term wind power forecasting. This study introduces a novel ultra‐short‐term multi‐step prediction model for wind power, which integrates the sparrow search algorithm (SSA), variational mode decomposition (VMD), gated recurrent unit (GRU), and support vector regression (SVR). An optimization variational mode decomposition technique is developed by adaptively determining VMD hyperparameters using SSA. The optimization VMD decomposes the original wind power sequence into sub‐modes, and the resulting sequence of decomposed sub‐modes calculates permutation entropy (PE) values. Sub‐modes with similar PE values are combined, reorganized, and categorized into high‐frequency and low‐frequency. High‐frequency sub‐modes data with high complexity and non‐stationarity are predicted by the GRU neural network. Low‐frequency sub‐modes data with low complexity and strong nonlinearity are predicted with SVR. The proposed model was evaluated against seven others using three error metrics: MAE, RMSE, and R2, along with their corresponding enhancement percentages. Experimental results indicate that the proposed model extracts detailed and trend information from the wind power series more effectively and stably than the comparison models. It also demonstrates superior multi‐step prediction performance, offering significant value for practical engineering applications.

Dammarane-Type Saponins from the Leaves of Vietnamese Ginseng (Panax vietnamensis Ha & Grushv.) and Their Acetylcholinesterase Inhibition In Vitro and In Silico.

Vietnamese ginseng (Panax vietnamensis Ha & Grushv.) represents one of the famous Panax spp. for valuable applications in both traditional and modern medicine; in which, its rhizome part has mainly been used as the medicinal materials based on the bioactive ginsenosides such as ginsenoside Rb1, ginsenoside Rg1, ginsenoside Rd, and majonoside R2. In modern medicine, the development of medicinal materials and utilization of medicinal plants are crucially based on standard bioactive ingredients, so this study to evaluate the leaves of Vietnamese ginseng as source of bioactive ginsenoside led to the identification of seven ginsenosides (1-7). Of them, ginsenoside Rd (2) and pseudoginsenoside RS1 (5) showed inhibitory effects on acetylcholinesterase in vitro with the IC50 values of 47.13 and 79.58 μM and supported by molecular docking analysis, in which ginsenoside Rd (2) and pseudoginsenoside RS1 (5) could play as allosteric inhibitors with high binding affinity (-8.5 and -9.4 kcal/mol) as evidenced by hydrogen bonding and hydrophobic interactions. The findings provided the scientific evidence for using the leaves of Vietnamese ginseng as an alternative source to the roots to enhance memory in traditional medicine as well as for further research on the anti-dementia effects of 2 and 5.

Equilibrium oxygen isotope fractionation in Mg(OH)2-Ca(OH)2-H2O: Implication from in situ high-temperature and high-pressure vibrational spectra

Abstract Equilibrium oxygen isotope fractionations have been extensively studied between H2O and minerals (like brucite-type hydroxides) in isotope geochemistry. In this study, pure 18O-enriched hydroxides were synthesized through reactions between M3N2 (M = Mg and Ca) powders and H218O water. In situ high-temperature (T) and high-pressure (P) Raman and Fourier transform infrared (FTIR) spectra were systematically collected for these phases. The observed 18O-16O effect on the frequency shifts aligns with the theoretical model, and has little impact on the anharmonic O-H potential well. The calculated intrinsic anharmonic parameters (ai) for both lattice and OH-stretching modes are essentially identical between M(16OH)2 and M(18OH)2, satisfying the classical limit for the Helmholtz free energy at extremely high temperature. Based on the measured vibrational data, the equilibrium oxygen isotope fractionation β(T,P) factors were modeled for both hydroxides. Both the intrinsic and external anharmonic contributions are smaller than the experimental uncertainties, and this model is also validated by ab initio calculation. Next, the 103·lnα(T) profiles were computed between brucite/portlandite and H2O, which are consistent with the reported oxygen isotope exchange measurements above 200 °C. Additionally, the discrepancy between this equilibrium model and the precipitation experiment below 120 °C also provides useful information for investigating the mechanism of kinetic isotope fractionation. The predicted 103·lnαbrucite-portlandite(T) curve also agrees well with points inferred from the 18O-16O exchange measurements, while the pressure effect can be ignored for 18O-16O fractionations. Therefore, vibrational spectroscopic measurements, including both Raman and infrared spectroscopies, have valuable applications in studying equilibrium non-metallic isotope fractionations in minerals.

Construction and validation of a clinical prediction model for sepsis using peripheral perfusion index to predict in-hospital and 28-day mortality risk.

Sepsis is a clinical syndrome caused by infection, leading to organ dysfunction due to a dysregulated host response. In recent years, its high mortality rate has made it a significant cause of death and disability worldwide. The pathophysiological process of sepsis is related to the body's dysregulated response to infection, with microcirculatory changes serving as early warning signals that guide clinical treatment. The Peripheral Perfusion Index (PI), as an indicator of peripheral microcirculation, can effectively evaluate patient prognosis. This study aims to develop two new prediction models using PI and other common clinical indicators to assess the mortality risk of sepsis patients during hospitalization and within 28 days post-ICU admission. This retrospective study analyzed data from sepsis patients treated in the Intensive Care Unit of Peking Union Medical College Hospital between December 2019 and June 2023, ultimately including 645 patients. LASSO regression and logistic regression analyses were used to select predictive factors from 35 clinical indicators, and two clinical prediction models were constructed to predict in-hospital mortality and 28-day mortality. The models' performance was then evaluated using ROC curve, calibration curve, and decision curve analyses. The two prediction models performed excellently in distinguishing patient mortality risk. The AUC for the in-hospital mortality prediction model was 0.82 in the training set and 0.73 in the validation set; for the 28-day mortality prediction model, the AUC was 0.79 in the training set and 0.73 in the validation set. The calibration curves closely aligned with the ideal line, indicating consistency between predicted and actual outcomes. Decision curve analysis also demonstrated high net benefits for the clinical utility of both models. The study shows that these two prediction models not only perform excellently statistically but also hold high practical value in clinical applications. The models can help physicians accurately assess the mortality risk of sepsis patients, providing a scientific basis for personalized treatment.

Electrochemical Investigation of Symmetric Aminoquinones

Abstract Quinones have a wide range of valuable properties and potential applications in medicinal chemistry, materials science, optoelectronic devices, and batteries. Molecular redesign using different functional groups, like amines, can optimize their properties and prevent undesired side reactions. However, the synthesis of aminoquinones can be particularly challenging at times, and there is a need for simple and efficient routes to access these compounds without metal catalysts or halogenated starting materials. Here, we demonstrate the synthesis and electrochemical characterization of a series of aminoquinones derived from renewable sources, namely vanillin or 2-methoxyhydroquinone. We employ a series of primary and secondary amines, varying in their electronic situation as well as steric demand. Depending on the type of starting material, either the desired aminoquinone or the related Schiff-base adduct was obtained. The aminoquinones were further explored towards their stability at different pH values. At extreme pH values, the deeply colored aminoquinones decompose, accompanied by decolorization of the solutions within a few minutes (pH 14) or hours (pH 1). At intermediate pH values (3-8) the aminoquinones are stable upon storage in solution, where they feature a quasi-reversible redox chemistry and fast, diffusion limited kinetics.

Enhancing domain-specific text generation for power grid maintenance with P2FT.

The digitization of operation and maintenance in the intelligent power grid equipment relies on a diverse array of information for smart decision-making. In the domain of intelligent decision generation, proficiency is contingent upon extensive learning from copious amounts of text. This necessitates not only robust processing capabilities but also a high level of specialization. In addressing situations where authorization is lacking, pre-trained language models (PLMs) have already provided ideas when confronted with specialized domains or tasks. In consideration of the complexity of textual content in the field of the power grid, which encompasses a multitude of specialized knowledge and involves an abundance of proprietary terminology, we have undertaken an exploration of pre-trained model specialization using the power grid domain as an example, specifically for the task of generating maintenance strategies. A two-stage fine-tuning approach (P2FT) is employed, utilizing a large-scale pre-training model specifically designed for natural language processing. The efficacy and practical value of this method were evaluated through multiple metrics, juxtaposed with other advanced approaches involving low-parameter or parameter-free fine-tuning methods. Through a meticulous analysis and validation of experimental outcomes, we have corroborated the feasibility and practical application value of employing this approach for pre-trained model specialization. Additionally, it has furnished valuable guidance for text generation within both the Chinese language domain and the power grid domain.

Atom-Vacancy-Defect-Derived Electric Hysteresis Loops and Stochastic Low-Frequency Noises in Few-Atom Layer MoS2.

Atom-vacancy-defects present in various materials yield numerous interesting physical phenomena, even obstructing high performance in some cases. On the other hand, their valuable applications to novel devices, such as nitrogen vacancy centers in diamond for quantum bits, have gathered significant attention. In particular, these tendencies become more substantial in two-dimensional (2D) (atomically) thin van der Waals layers. However, correlations with various kinds of atom defects are still under exploration. Herein, we find the stochastic behaviors of large hysteresis loops with strong photoresponse in the static electrical properties in few-atom layer semiconductors, molybdenum disulfide (MoS2). The temperature dependence and transmission electron microscopy reveal that they arise from pairs of two neighboring in-plane S-vacancy defects, which predominantly present only around the interface at the MoS2 flake/substrate, with activation energies ∼0.35 eV. The low-frequency (f) (LF) noise measurements clarify a high f shift in the two 1/f2-dependent regimes, implying stochastic behaviors of electric charges through the S-vacancy pairs with high-speed charge(spin) transitions across low kinetic energy barriers between narrow discrete states. The shallow energy sates are formed from the highly uniform S-vacancy pairs interacting with Mo atoms, which act like quantum dots. The observed stochastic operation holds promise for various application, particularly for probabilistic neuromorphic computation in artificial intelligence.

Research on hydroacoustic signal processing algorithm based on B-spline and Hilbert transform

Purpose This paper aims to solve the problems of baseline drift, susceptibility to abnormal data interference during baseline drift processing, and phase inconsistency in underwater acoustic target detection and signal processing of single microelectromechanical systems (MEMS) vector hydrophone. To this end, this paper proposes a baseline drift removal algorithm based on Huber regression model with B-spline interpolation (H-BS) and a phase compensation algorithm based on the Hilbert transform. Design/methodology/approach First, the Huber regression model is innovatively introduced into the conventional B-spline interpolation (B-spline) to solve the control point vectors more accurately and to improve the anti-interference ability of the abnormal data when the B-spline interpolation fitting removes baseline drift and the baseline drift components in the signals are fitted accurately and removed by the above method. Next, the Hilbert transform is applied to the three-channel output signals of the MEMS vector hydrophone to calculate the instantaneous phase and the phase compensation is performed on the vector X/Y signals with the scalar P signal as the reference. Findings Through simulation experiments, it is found that H-BS proposed in this paper has smaller processing error and better robustness than variational modal decomposition and B-spline, but the H-BS algorithm takes slightly longer than the B-spline. In the actual lake test experiments, the H-BS algorithm can effectively remove the baseline drift component in the original signal and restore the signal waveform, and after the Hilbert transform phase compensation, the direction of arrival estimation accuracy of the signal is improved by 1°∼2°, which realizes high-precision orientation to the acoustic source target. Originality/value In this paper, the Huber regression model is introduced into B-spline interpolation fitting for the first time and applied in the specialized field of hydroacoustic signal baseline drift removal. Meanwhile, the Hilbert transform is applied to phase compensation of hydroacoustic signals. After simulation and practical experiments, these two methods are verified to be effective in processing hydroacoustic signals and perform better than similar methods. This study provides a new research direction for the signal processing of MEMS vector hydrophone, which has important practical engineering application value.

Analyze the moment of inertia of an object in uniformly accelerated linear motion and angular motion

Abstract. In studying particle kinematics, calculus plays an important role in analyzing and solving motion problems as a key mathematical tool. This paper mainly discusses the application of calculus in uniformly accelerated linear motion (LM) and angular motion (AM), focusing on the relationship between position, velocity, and acceleration and the analysis of the rate of change of motion. Through the study of uniformly accelerated LM, this paper guides researchers to use calculus to establish the mathematical relationship between position, velocity, and acceleration and verifies the accuracy of these formulas through practical cases. Similarly, in AM, this paper provides an in-depth understanding of the dynamic behavior of an object rotating around a fixed axis through calculus and calculates and applies important parameters in AM to solve practical problems by analyzing angular velocity and acceleration. In addition, this paper demonstrates the potential of calculus for a wide range of applications in dealing with variable speed motion and rate of change analysis. By integrating the rate of change of displacement and acceleration, it provides a precise tool for the optimization and resolution of engineering problems. In conclusion, calculus not only enriches the theory of kinematics but also demonstrates its important value in practical applications.

Identification of material excavation difficulty and uncertainty analysis based on Bayesian deep learning

Accurately assessing the difficulty of material excavation is crucial for reducing excavator energy consumption, ensuring operational safety, and optimizing excavator efficiency. Addressing the challenges of uncertain and difficult-to-judge excavation conditions for underground materials, this paper proposes a Bayesian deep learning-based method that integrates excavation process data to identify excavation difficulty. Firstly, we constructed a deep learning model based on Bayesian theory and decomposed the uncertainty of the identification results into aleatory uncertainty and epistemic uncertainty. Next, through a mechanistic analysis of the interaction between materials and the excavator bucket during excavation, we identified the input features for the model. Finally, we validated the effectiveness of the method through experiments. The results show that the proposed method not only accurately identifies the excavation difficulty of the material but also quantifies and decomposes the uncertainty of the identification results, demonstrating both theoretical significance and practical application value.

Numerical simulation of the dual-loop different velocity air curtain system in the operating room

The problem of cross-infection between patients and surgeons in the operating room had been widely concerned by the international community. The purpose of this study had been to explore the potential of a Dual-loop Different Velocity Air Curtain System (DDVACS) being applied in the operating room to reduce cross-infection. In this study, the performance of the system was studied using the computational fluid dynamics (CFD) numerical simulation method, with the operating conditions of different air supply velocities inside and outside being set. The RNG k-ε model, Discrete Phase Model (DPM), and Species Transport model were used to simulate the air distribution, particle diffusion, and virus diffusion. The main research results showed that, when the DDVACS was operated at an inner air supply velocity of 0.24 m/s and an outer air supply velocity of 0.45 m/s, the virus concentration in the surgical area had decreased by 89%, the particulate matter concentration in the surgical area had decreased by 78.08%, but the deposition of particulate matter had increased by 21.62%. This study concluded that DDVACS had been found to effectively reduce the concentration of cross-infection. Finally, the velocity difference between the inner and outer sides was controlled within 2 times. This setting had been arranged to make the DDVACS work better. This study's contribution was to further verify the application of the DDVACS in operating rooms, providing a new perspective and empirical support for related fields and possessing certain practical application value.

Analysis of Candidate miRNAs' Expression in Pancreatic Cancer.

Pancreatic cancer (PC) is one of the most aggressive types of cancer. Despite advances in molecular diagnostics, PC diagnosis relies on imaging technologies and morphological assessment of fine needle aspirates (FNAs). MicroRNA (miRNA) involvement in PC pathogenesis and potential diagnostics application have been suggested, albeit current supporting evidence is lacking. Here, we investigated the association of selected miRNAs with PC incidence and clinical characteristics. Fold expression of miR-216a-3p, -217-5p, -221-3p, -222-3p, and miR-196a-5p was assessed in 73 PC FNA cell-block sections and 6 healthy pancreas tissues using Taqman advanced miRNA assays. Potential miRNA targets were ascertained using immunocytochemistry. miR-196a-5p was upregulated in PC compared to healthy pancreatic tissue (β = -0.05, 95% CI: -0.065 - (-0.035); p < 0.001). miR-221-3p and miR-222-3p fold expression were strongly correlated (r = 0.897, p < 0.001), whereas miR-196a-5p fold expression correlated with that of miR-221-3p (r = 0.688, p < 0.001) and miR-222-3p (r = 0.489, p < 0.001). Moreover, miR-196a-5p fold expression positively correlated with tumor stage (r = 0.32, p = 0.034). miR-217-5p fold expression inversely correlated with KRAS expression (r = -0.69, p = 0.0027). Our study shows miR-196a-5p has reasonable specificity to PC and thus may have diagnostic and prognostic potential in PC as proposed in the literature. Moreover, KRAS and NFKBIA may be potential targets for miR-217-5p and miR-196a-5p, respectively. Thus, these miRNAs may be involved in tumor progression and may have valuable applications in novel therapeutics or treatment monitoring.

Valorisation of hop leaves for their bioactive compounds: Identification and quantification of phenolics across different varieties, crop years and stages of development.

Hop leaves, a by-product from hop cone harvesting, contain phenolic compounds of potential value for food or beverage applications. However, the abundant phenolics in hop leaves remain largely unquantified. This study quantified phenolics in hop leaves over two crop years, for three commercially significant varieties, at different developmental stages post-flowering. Ethanolic hop extracts were characterised using LC-ESI-qTOF-MS/MS and HPLC-DAD for the annotation and quantification of phenolics and bitter resins. Hop leaf phenolic profile exhibited considerable structural diversity, differing significantly from that of respective cones. Kaempferol/quercetin 3-O-glycosides and chlorogenic acids were the most abundant sub-groups with phenolic acids, procyanidins, prenylflavonoids and bitter resins also present. Phenolic profile was mainly variety-dependent with some crop year and developmental effects. Flavonol 3-O-glycosides were the main compounds driving varietal differences. Findings demonstrate the structural diversity and high concentrations of phenolic compounds in hop leaf extracts and their potential as a source of bioactives for valorisation.

Colour guided ground-to-UAV fire segmentation

Leveraging ground-annotated data for scene analysis on unmanned aerial vehicles (UAVs) can lead to valuable real-world applications. However, existing unsupervised domain adaptive (UDA) methods primarily focus on domain confusion, which raises conflicts among training data if there is a huge domain shift caused by variations in observation perspectives or locations. To illustrate this problem, we present a ground-to-UAV fire segmentation method as a novel benchmark to verify typical UDA methods, and propose an effective framework, Colour-Mix, to boost the performance of the segmentation method equivalent to the fully supervised level. First, we identify domain-invariant fire features by deriving fire-discriminating components (u*VS) defined in colour spaces Lu*v*, YUV, and HSV. Notably, we devise criteria to combine components that are beneficial for integrating colour signals into deep-learning training, thus significantly improving the generalisation abilities of the framework without resorting to UDA techniques. Second, we perform class-specific mixing to eliminate irrelevant background content on the ground scenario and enrich annotated fire samples for the UAV imagery. Third, we propose to disentangle the feature encoding for different domains and use class-specific mixing as robust training signals for the target domain. The framework is validated on the drone-captured dataset, Flame, by using the combined ground-level source datasets, Street Fire and Corsica Wildfires. The code is available at https://github.com/Rui-Zhou-2/Colour-Mix.

Sparse discriminant manifold projections for automatic depression recognition

In recent years, depression has become an increasingly serious problem globally. Previous research have shown that EEG-based depression recognition is a promising technique to serve as auxiliary diagnosis methods that provide assistance to clinicians. Typically, in clinical studies, due to the multichannel nature of EEG, the extracted features usually are high-dimensional and contain many redundant information. Therefore, it is necessary to perform dimensionality reduction before classification to improve the performance of machine learning algorithms. However,existing dimensionality reduction techniques do not design the objective function based on the characteristics of EEG signal and the goal of depression recognition, so they are less suitable for dimensionality reduction of EEG features. To solve this problem, in this paper we propose a novel dimensionality reduction technique called sparse discriminant manifold projections(SDMP) for depression recognition. Specifically, the use of the ℓ2-norm instead of the squared ℓ2-norm as a similarity measure in the objective function reduces sensitivity to noise and outliers. Moreover, the local geometric structure and global discriminative properties of data are integrated, which makes the extracted features more discriminative. Finally, the ℓ2,1-norm regularization is introduced to achieve feature selection. Furthermore, The formulation is extended to the ℓ2,p-norm regularization case, which is more likely to offer better sparsity when 0<p<1. Extensive experiments on EEG data show that the SDMP achieves the competitive performance compared with other state-of-the-art dimensionality reduction methods. It also shows the practical application value of our method in detecting depression.

An automatic hybrid mesh generation method for combustion fluid dynamics simulations

PurposeAutomatic mesh generation is still a challenge problem for combustion fluid dynamics simulations because of the high-quality requirement and complexity of geometries. This paper aims to find an efficient automatic analysis model creation and mesh generation method to save the time for pre-processing in numerical simulations.Design/methodology/approachBased on the previous work, we explore effective model processing and mesh generation methods from practical engineering applications. Considering the automation and high quality requirement, we construct an automatic mesh generation procedure for combustion fluid dynamics problems.FindingsThe numerical results show that the procedure we proposed in this paper can automatically generate high quality mesh for combustion fluid dynamics simulations. The strategy of fluid model construction is time-saving and with high-precision. The mesh generation method in our procedure is automatic and efficient.Practical implicationsThe procedure proposed in this paper is applicable to the practical engineering application model, such as aircraft simulation, aeroengine simulation and so on. The procedure has been integrated into an numerical simulation software.Originality/valueThe method proposed in this paper has very practical application value. It can be used in the practical application and saves a lot of manual processing time for numerical stimulation specialists.

Border Trap-Enhanced Ga2O3 Nonvolatile Optoelectronic Memory.

Nonvolatile deep ultraviolet optoelectronic memory (DUVOEM) holds immense potential in cyberphysical systems, offering high storage density, swift conversion speeds, and robust data security. However, the current data retention time, typically limited to milliseconds or hours, mostly underperforms the expectations of years as a nonvolatile memory. In this work, we present a β-Ga2O3/SiO2/Si thin-film transistor DUVOEM with an enhanced data storage capability via trapping and releasing of photogenerated holes in border traps. Specifically, the photogenerated holes in β-Ga2O3 will tunnel through SiO2 and be captured by these defects. Innovatively, the much slower holes' release process from the border traps has been harnessed in developing outstanding nonvolatile optoelectronic memories. Rapid writing and erasing speeds, long-time retention (≥10 years), and high robustness demonstrate its practical application values. This study not only provides a novel strategy for nonvolatile DUVOEM but also provides an instance of functionalizing β-Ga2O3 memory with common defects in Si technology.

Traffic flow parameter estimation for complex environments based on improved YOLOv8

Estimation of traffic flow parameters based on computer vision is still a very popular challenge. In particular, the detection problems such as occlusion, small targets, luminance variations, and lighting jitter, etc., are caused by dense traffic scenes and complex weather conditions. Previous studies about detecting and tracking methods focused on daytime and nighttime or single weather condition, but the performance of this model will decline dramatically due to complex conditions. In this paper, a framework for estimating traffic flow parameters in complex environments is presented, in which Ghost-YOLOv8 vehicle detection model is proposed. GhostConv is used to replace part of Conv and a new C2fGhost module is designed to replace part of C2f, reducing the number of parameters and improving the detection performance of the model; The global attention mechanism module is added to the neck network, which strengthens the semantic and location information in the features and improves the model’s ability of feature fusion; In view of the loss of semantic information caused by different scales in detecting small targets, a small target detection layer is added to enhance a combination of deep and shallow semantic information; The GIoU bounding loss function is used instead of the original loss function, which improves the performance of the network’s bounding box regression. Then, combining DeepSORT tracking algorithm and virtual line counting method, the proposed framework estimates traffic flow parameters, including volume, speed, and density in complex environments. The experimental results show that compared with the original model in the UA-DETRAC dataset, the improved YOLOv8 model’s the accuracy is increased by 1.4%, and the parameter number and model size are reduced by 0.229G and 0.2MB respectively. In addition, the framework has good robustness, reaching 97.56% accuracy when estimating average traffic flow parameters. It is sufficient to overcome complex weather conditions and effectively help control traffic for intelligent transportation systems and traffic management to provide good data support. In conclusion, it shows that the model can reduce the number and size of model parameters and improve the detection precision as well as meeting the requirements of edge computing devices and having better real-time performance, so it has practical application value.