Traditional Processing Methods Research Articles

Entity Extraction of Key Elements in 110 Police Reports Based on Large Language Models

With the rapid advancement of Internet technology and the increasing volume of police reports, relying solely on extensive human labor and traditional natural language processing methods for key element extraction has become impractical. Applying advanced technologies such as large language models to improve the effectiveness of police report extraction has become an inevitable trend in the field of police data analysis. This study addresses the characteristics of Chinese police reports and the need to extract key elements by employing large language models specific to the public security domain for entity extraction. Several lightweight (6/7b) open-source large language models were tested as base models. To enhance model performance, LoRA fine-tuning was employed, combined with data engineering approaches. A zero-shot data augmentation method based on ChatGPT and prompt engineering techniques tailored for police reports were proposed to further improve model performance. The key police report data from a certain city in 2019 were used as a sample for testing. Compared to the base models, prompt engineering improved the F1 score by approximately 3%, while fine-tuning led to an increase of 10–50% in the F1 score. After fine-tuning and comparing different base models, the Baichuan model demonstrated the best overall performance in extracting key elements from police reports. Using the data augmentation method to double the data size resulted in an additional 4% increase in the F1 score, achieving optimal model performance. Compared to the fine-tuned universal information extraction (UIE) large language model, the police report entity extraction model constructed in this study improved the F1 score for each element by approximately 5%, with a 42% improvement in the F1 score for the “organization” element. Finally, ChatGPT was employed to align the extracted entities, resulting in a high-quality entity extraction outcome.

Fusing binocular vision and deep learning to detect dynamic wheel-rail displacement of high-speed trains

Large displacements between the wheel and rail lead to track damage or even train derailment. Detecting dynamic wheel-rail displacement (DWRD) can provide a technical means for discriminating the operational health status of rail vehicles and formulating active control strategies. In this paper, a new method, which fuses binocular vision (BV) and deep learning (DL), is proposed to detect the DWRD online, the method is abbreviated as BV-DL. First, a binocular camera is used to capture the dynamic wheel-rail contact video online, while two lasers are equipped to focus on the wheel-rail contact region to locate some keypoints in the wheel-rail contact images. Then, an ROI (region of interest) detection network is proposed to recognize the wheel-rail contact region automatically, and a keypoint detection network is proposed to detect the wheel-rail pixel displacement (WRPD). Finally, a 3D coordinate transformation (3DCT) model is built to calculate the wheel-rail world displacement in real-time, i.e., the DWRD between the keypoints. To test the performance of the method, a laser displacement sensor-based method is used to verify the reliability of the 3DCT model. Meanwhile, a traditional image processing method, a MobileNet V3-based method, and a YOLOX-s-based method are introduced to predict the WRPD. Experimental results show that compared with the four methods, the proposed BV-DL method can achieve optimal results in pixel coordinate calculation and 3D coordinate computation.

Traditional Processing Can Enhance the Medicinal Effects of Polygonatum cyrtonema by Inducing Significant Chemical Changes in the Functional Components in Its Rhizomes.

The aims of this study were to explore the significant chemical changes in functional components induced by the traditional processing method and evaluate whether this method based on nine cycles of steaming and drying can effectively enhance the medicinal effects of Polygonatum cyrtonema rhizome. A global analysis on dynamic changes in secondary metabolites during nine processing cycles was performed, and the significantly differentially accumulated secondary metabolites were initially identified based on the secondary metabolome. Unsupervised principal component analysis (PCA), hierarchical clustering analysis (HCA), and orthogonal partial least squares discriminant analysis (OPLA-DA) on secondary metabolites clearly showed that processing significantly increased the global accumulation of secondary metabolites in processed P. cyrtonema rhizomes compared to unprocessed crude rhizomes. The first six processing cycles induced drastic changes in the accumulation of functional components, while the last three did not induce further changes. The accumulations of most functional components were significantly enhanced after the first three cycles and stabilized after six cycles; meanwhile, the first three cycles also led to numerous new components. However, the enhancing effects were unavoidably reversed or weakened under continued processing lasting 6-9 cycles. Furthermore, continued processing also reduced the contents of a small number of original components to undetectable levels. Processing induced some significantly enriched Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways, among which the first three processing cycles enhanced the synthesis of various secondary metabolites and significantly affected the metabolisms of amino acids. In conclusion, this study not only reveals that processing can effectively enhance the medicinal effects, by two main mechanisms including enhancing chemical synthesis and inducing structural transformation of functional components, but also provides theoretical guidance for the optimization of the traditional processing method based on nine cycles of steaming and drying for achieving optimal effects on enhancing the medicinal effects of P. cyrtonema rhizome.

A model is worth tens of thousands of examples for estimation and thousands for classification

Traditional signal processing methods relying on mathematical data generation models have been cast aside in favour of deep neural networks, which require vast amounts of data. Since the theoretical sample complexity is nearly impossible to evaluate, these amounts of examples are usually estimated with crude rules of thumb. However, these rules only suggest when the networks should work, but do not relate to the traditional methods. In particular, an interesting question is: how much data is required for neural networks to be on par or outperform, if possible, the traditional model-based methods? In this work, we empirically investigate this question in three simple examples covering estimation and classification, where the data is generated according to precisely defined mathematical models, and where well-understood optimal or state-of-the-art mathematical data-agnostic solutions are known. A first problem is deconvolving one-dimensional Gaussian signals, a second one is estimating a circle’s radius and location in random grayscale images of disks, and a third one both classifies the presence of a line and locates it when present in a binary random dot image. By training various networks, either naive custom designed or well-established ones, with various amounts of training data, we find that networks require tens of thousands of examples for estimation in comparison to the traditional methods and thousands for classification, whether the networks are trained from scratch or even with transfer-learning or finetuning.

Fabrication of controllable porous tungsten tips for indium FEEP by dynamic reciprocating electrochemical etching combined with ultrasonic cleaning method

The porous tungsten emitter is a core component of field emission electric propulsion (FEEP). It combines the advantages of sharp tip and internal propellant transport. The emission performance of a FEEP thruster is highly depended on the emitter's tip apex radius. Traditional processing methods for solid tungsten tips are unsuitable for porous tungsten due to its unique structure and electrochemical properties. This paper introduces a novel approach combining dynamic reciprocating electrochemical etching with ultrasonic cleaning to fabricate cone-shaped porous tungsten tips with controlled apex radii. By alternating between 5 cycles of etching and an ultrasonic cleaning, a 1.9 μm apex radius is achieved after 90 repetitions. The etching current waveform reveals the material dissolution mechanism. The electrochemical characteristics of porous tungsten are investigated, with an optimal processing voltage of 2 V identified. To refine the apex radius and prevent tip collapse, a 1.9 V etching voltage is used in the final stages, maintaining the apex within 2 μm. The effectiveness of the fabricated tip is confirmed through wetting in indium and ignition tests.

Water Conservancy Hub Geometric Modeling Based on Point Clouds

For hydraulic engineering, accurate geometric modeling is crucial for design, analysis, and maintenance. In recent years, point cloud technology, with its high precision and detailed data capture capabilities, has gradually become the preferred method for geometric modeling in hydraulic engineering. However, traditional point cloud processing methods exhibit slow data processing speeds and poor mesh quality when reconstructing 3D models. Therefore, we have studied the meshing method for the 3D reconstruction of large-scale discrete point cloud data and proposed a meshing method based on the Delaunay criterion and the growth method to achieve rapid meshing of discrete point cloud data. To demonstrate the efficiency and convenience of the proposed method, point cloud reconstruction experiments are conducted on various buildings. We compare our approach with the Poisson, BPA, and ONet methods. Additionally, we apply it to the Jiangsu Gaogang Water Conservancy Hub in China. The results show that our proposed method is superior in both reconstruction quality and efficiency, providing a promising solution for the reconstruction of 3D building models from point clouds.

Review of Hybrid Denoising Approaches in Face Recognition

Statistically, image denoising is one of the key pillars of image processing and picture acquisition, which also is utilized to clear the noisy images. Over the last years, there is an increase of study subjects that are devoting to designing and making noise cancellation methods. This study reviews all major image denoising techniques, with a special emphasis on integrated deep learning approaches as well as traditional signal processing methods. The review presents a broad array of techniques for instance convolutional neural networks (CNNs), wavelet transforms, hybrid models, and their emendations. The lecturer will focus on the advantages, as well as the disadvantages, of each methodology along with their appropriateness in various fields, from which the current state of the art image denoising can be concluded. On the other hand, the paper discusses critical barriers leading to further prospects of research in cybersecurity and cybercrime prevention This review is important in that it aims to serve researchers, practitioners, and enthusiasts who would like to peer into the new trends and developments in denoise image generation.

Освоение моделирования как элемента подготовки студентов-гуманитариев в условиях развития технологии больших данных

g Data is a qualitatively new information object that requires special methods of analysis and processing. The peculiarity of Big Data is the following: traditional methods of data processing in the case of Big Data may not be effective, since they require very large time frames and computing resources; Big Data, in many ways, loses its discrete structure and approaches its properties to structures of continuity, which complicates the use of traditional methods of discrete analysis. The article discusses approaches to the study of Big Data in humanitarian universities. The sources of the emergence of big data in this area are discussed as an imbalance between the syntactic and semantic aspects of the information object. It is substantiated that in humanitarian universities it is advisable to place emphasis on mastering the modeling methodology, which, in many respects, allows eliminating this imbalance.

Tracking Metabolite Variations during the Degradation of Vegetables in Rice Bran Bed with Intact-State Nuclear Magnetic Resonance Spectroscopy.

Fermentation-a process of compound degradation by microorganisms-is a traditional food processing method utilized worldwide for the long-term preservation of fresh foods. In recent years, fermented foods have gained attention as health foods. Fermentation increases the nutritional value of ingredients, producing complex flavors and aromas. To identify unknown components in fermented foods, it is necessary to analyze compounds and conditions nondestructively and comprehensively. We performed intact-state nuclear magnetic resonance (NMR) spectroscopy using intermolecular single quantum coherence (iSQC) to detect the degradation of vegetables directly and nondestructively. We used two types of vegetables and a rice bran bed (nukazuke), which is used for traditional vegetable fermentation in Japan. Major metabolites such as saccharides, organic acids, and amino acids were identified in iSQC-sliced spectra. Comparing NMR signal intensities during degradation revealed the transition of metabolites characteristic of lactic acid fermentation. A pathway-based network analysis showed pathways involved in amino acid metabolism and lactic acid fermentation. Our analytical approach with intact-state NMR spectroscopy using iSQC demonstrated that it may be effective in other experimental systems, allowing for the evaluation of phenomena that have been conventionally overlooked in their true state.

3D-printed polymer-derived ceramics with tunable cellular architectures

Ceramic materials possess high mechanical strength and environmental stability, but their brittleness limits their suitability for structural applications. A solution lies in using polymer-derived ceramics (PDCs), which offer enhanced toughness and versatility in shaping unlike traditional ceramic processing methods. This study explores tunable ceramic cellular architectures based on triply periodic minimal surface (TPMS) designs, fabricated via stereolithography (SLA) using a silicon oxycarbide precursor formulated for vat photopolymerization. By combining the preceramic polymer with a photoinitiator, crosslinkers, and other additives, intricate shapes are 3D-printed and then pyrolyzed under nitrogen, resulting in PDCs with complex TPMS geometries. The toughness, strength, and stiffness of the 3D-printed structures are evaluated through quasi-static compression experiments. Comprehensive material and microstructural characterizations of the PDCs are performed pre- and post-pyrolysis, employing visual inspection, X-ray micro-tomography, thermogravimetric analysis, energy dispersive X-ray spectroscopy, density, and rheological measurements. Optimization of 3D printing and pyrolysis parameters yields ceramic structures with 2.2 MPa compressive strength and 330 MPa stiffness with a lattice density of 0.5 g cm-3. The ceramic material, including porosity, had a maximum density of 1.63 ± 0.01 g cm-3. This low-cost SLA 3D printing technique is ideal for creating thin features and customized structures of bio-inspired, architectured ceramics. Furthermore, the process exhibits excellent printability, being compatible with common and cost-effective SLA, DLP, and LCD 3D printers.

Recent Advances in Non-Contact Food Decontamination Technologies for Removing Mycotoxins and Fungal Contaminants.

Agricultural food commodities are highly susceptible to contamination by fungi and mycotoxins, which cause great economic losses and threaten public health. New technologies such as gamma ray irradiation, ultraviolet radiation, electron beam irradiation, microwave irradiation, pulsed light, pulsed electric fields, plasma, ozone, etc. can solve the problem of fungal and mycotoxin contamination which cannot be effectively solved by traditional food processing methods. This paper summarizes recent advancements in emerging food decontamination technologies used to control various fungi and their associated toxin contamination in food. It discusses the problems and challenges faced by the various methods currently used to control mycotoxins, looks forward to the new trends in the development of mycotoxin degradation methods in the future food industry, and proposes new research directions.

Obtention of Suitable Pregnant Leach Solution (PLS) for Copper Solvent Extraction Plants from Copper Concentrate Using Hydrogen Peroxide and Iodine in a Sulfuric Acid–Chloride Medium

Copper leaching presents an environmentally friendly alternative to traditional sulfide ore processing methods. This study investigates an efficient leaching process for copper concentrate, utilizing a solution of sulfuric acid (H2SO4) and potassium iodide (KI) in a chloride medium (NaCl), enhanced by hydrogen peroxide (H2O2) at room temperature. A significant aspect of this research was optimizing the KI concentration to minimize iodide sublimation into iodine gas (I2). Through the experimental design, the optimal dosages of reagents were determined, leading to maximized copper extraction of approximately 27% in 45 min of testing at room temperature. The results showed that it is possible to obtain a suitable pregnant leach solution (PLS) (i.e., in the range of 3 to 8 g/L of Cu) for treatment in available copper solvent extraction (SX) plants with a cost of less than 4.5 USD/t Cu, according to the economic analysis carried out. The results of this study determine the most effective operational conditions for leaching and ensure a suitable PLS for SX plants in a cost-effective and environmentally friendly manner. This approach could significantly contribute to more sustainable practices in the mining and processing of copper ores.

Digital Transformation Techniques for Manuscripts

The world is witnessing continuous technological developments that have an impact on societies and institutions. Therefore, all countries seek to keep pace with these developments and introduce them into various fields. Digital transformation has become one of the most important necessities that all institutions must adopt today, as digital transformation is capable of creating a competitive technological environment to achieve the highest levels of quality at the lowest costs. Manuscripts are of interest to many scholars and researchers due to their scientific and artistic value. Preserving them means preserving national identity, especially with the emergence of technology. It has become imperative to preserve manuscripts from damage and loss and make them available to the largest number of beneficiaries. In the past, Arabs were among the most important peoples who contributed to the preservation of manuscripts. One of the most important challenges facing modern society is managing, preserving, and retrieving information from the vast amount of paper documents and files that are increasing daily. By using technology, the time taken to search for the required document or distribute documents to the relevant parties leads to a significant increase in the productivity of the institution. Information technology and electronic communication systems have revolutionized traditional methods of processing, preserving, and exchanging information. The presence of advanced technologies provides us with ease and accuracy in accessing information. Archiving as a general concept relies on storing and classifying papers in a way that makes it easy to refer back to them when needed. With the advent of computers, the process has become easier, providing more flexibility and ease in retrieval, as well as reducing the costs of preservation and storage. This technology has a direct impact on the archival profession, as archivists have turned to electronic means and imaging storage systems, as well as audiovisual methods, in an attempt to solve the storage and preservation problem. Modern digital technology affects all areas of life due to the integrated scientific environment it provides. All societies have begun to transition towards the digital age because of its remarkable development in information and communication technology. Keywords: Information, Manuscript, Technology, Scanning, Digitization

Effects of different post-harvest processing methods on changes in the active ingredients of licorice based on LC-MS and plant metabolomics.

Licorice, the dried roots and rhizomes of the Glycyrrhiza uralensis Fisch., holds a prominent status in various formulations within the realm of Chinese medicinal practices. The traditional processing methods of licorice hinder quality assurance, thus prompting Chinese medicine researchers to focus on the fresh processing methods to enhancing processing efficiency and quality. This study aimed to identify the differential compounds of licorice between traditional and fresh processing methods and provide a scientific basis for the fresh processing of licorice and for further research on the processing mechanism. A methodology integrating ultra-performance liquid chromatography with quadrupole-time-of-flight tandem mass spectrometry combined with multivariate statistical analysis was employed to characterize the differential compounds present in licorice between traditional processing and fresh processing. The results derived from principal component analysis and heat map analyses underscored significant differences in the content of bioactive compounds between the two processing methods. By applying conditions of VIP > 1.5 and p < 0.05, a total of 38 differential compounds were identified through t tests, and the transformation mechanisms of select compounds were illustrated. The adoption of fresh processing techniques not only improved processing efficiency but also significantly enhanced the preservation of bioactive compounds within licorice. This research has established a rapid and efficient analytical method for the identification of differential compounds present in differently processed licorice products.

Lane Line and Object Detection Using Yolo v3

In the contemporary age, creating autonomous vehi- cles is a crucial starting point for the advancement of intelligent transportation systems that rely on sophisticated telecommu- nications network infrastructure, including the upcoming 6g networks. The paper discusses two significant issues, namely, lane detection and obstacle detection (such as road signs, traffic lights, vehicles ahead, etc.) using image processing algorithms. To address issues like low accuracy in traditional image processing methods and slow real-time performance of deep learning-based methods, barriers for smart traffic lane and object detection algorithms are proposed. We initially rectify the distorted image resulting from the camera and then employ a threshold algorithm for the lane detection algorithm. The image is obtained by extracting a specific region of interest and applying an inverse perspective transform to obtain a top-down view. Finally, we apply the sliding window technique to identify pixels that belong to each lane and modify it to fit a quadratic equation. The Yolo algorithm is well-suited for identifying various types of obstacles, making it a valuable tool for solving identification problems. Finally, we utilize real-time videos and a straightforward dataset to conduct simulations for the proposed algorithm. The simula- tion outcomes indicate that the accuracy of the proposed method for lane detection is 97.91% and the processing time is 0.0021 seconds. The proposal for detecting obstacles has an accuracy rate of 81.90% and takes only 0.022 seconds to process. Compared to the conventional image processing technique, the proposed method achieves an average accuracy of 89.90% and execution time of 0.024 seconds, demonstrating a robust capability against noise. The findings demonstrate that the suggested algorithm can be implemented in self-driving car systems, allowing for efficient and fast processing of the advanced network.

Comparative analysis of time-frequency transform-based differential protection strategy for distributed generation integrated microgrid

This study presents a differential protection strategy for distributed generation-integrated power systems. It explores and compares two enhanced traditional signal processing methods: the Hilbert-Huang Transform (HHT) and the S-transform. The HHT has been improved by incorporating Variational Mode Decomposition (VMD), while the S-transform has been augmented with a Fourier component to create the Fast Fourier Discrete Orthonormal Stockwell Transform (FFDOST). VMD and FFDOST are applied to three-phase current signals from microgrid buses to calculate the Spectral Energy (SE). Fault detection is based on Differential Energy (DE), which measures the difference in SE between two specific faulty buses, with a tripping signal activated when DE surpasses a predetermined threshold. The effectiveness of this approach is rigorously tested and validated on a typical DG-integrated microgrid that includes synchronous generators and wind farms. A comprehensive set of case studies conducted on an IEC microgrid model examines various fault types (symmetrical and asymmetrical) across different operational modes (islanded and grid-connected) and configurations (radial and looped) under varying noise conditions. The performance of these two modified differential strategies is critically compared to determine the more effective method. Detailed sample-wise analysis and graphical presentations enrich the documentation of results. The VMD-based method outperforms the FFDOST approach, detecting faults 3 to 15 samples sooner and achieving quicker fault detection by approximately 0.05 milliseconds. Comparative analyses suggest that VMD is a faster, more secure, and highly reliable method for fault detection and mitigation, enhancing microgrids' operational security and reliability.

Research on high-speed classification and location algorithm for logistics parcels based on a monocular camera

The rapid development of the logistics industry has driven innovations in parcel sorting technology, among which the swift and precise positioning and classification of parcels have become key to enhancing the performance of logistics systems. This study aims to address the limitations of traditional light curtain positioning methods in logistics sorting workshops amidst high-speed upgrades. This paper proposes a high-speed classification and location algorithm for logistics parcels utilizing a monocular camera. The algorithm combines traditional visual processing methods with an enhanced version of the lightweight YOLOv5 object detection algorithm, achieving high-speed, high-precision parcel positioning. Through the adjustment of the network structure and the incorporation of new feature extraction modules and ECIOU loss functions, the model’s robustness and detection accuracy have been significantly improved. Experimental results demonstrate that this model exhibits outstanding performance on a customized logistics parcel dataset, notably enhancing the model's parameter efficiency and computational speed, thereby offering an effective solution for industrial applications in high-speed logistics supply.

Mechanical property of powder bed fusion printed carbon fiber reinforced polyetheretherketone and enhanced interlayer strength by post‐processing

AbstractAdditive Manufacturing (AM) of carbon fiber reinforced polyetheretherketone (CF/PEEK) composites with excellent performance by Powder Bed Fusion (PBF) has great potential in the applications where high strength and high service temperature are required. However, the reinforcement mechanisms of CF/PEEK composites based on the PBF process are complex and show significant differences from traditional processing methods. In this paper, the mechanical properties of CF/PEEK prepared by PBF were investigated. The results showed that its X‐axis tensile strength and modulus were approximately equal to or a bit lower than the theoretical value with randomly distributed CF, but much lower than the theoretical value with unidirectional distribution of CF in composites. Meanwhile, mechanical strength anisotropy was observed. The composites' Y‐axis tensile strength was slightly lower than X‐axis, and the lowest strength existed in Z‐axis. To improve the weak Z‐axis strength, post‐processing of annealing and solvent‐induced crystallization were carried out, and the Z‐axis tensile strength of post‐processed parts reached 50 MPa, which was 62% higher than untreated samples. Consequently, suitable CF and proper post‐ processing for AM of high strength CF/PEEK were proposed, which would promote the industrial application of this technology.Highlights The effect of CF on mechanical property of CF/PEEK prepared by PBF were studied. The optimal CF for PBF of high strength CF/PEEK were obtained. The post‐processing method for improving the Z‐axis strength were proposed. CF/PEEK with comprehensive mechanical property was successfully prepared by PBF.

Segmentation Method of Concrete Small Cracks Based on UAV Images

Introduction:: Cracks are one of the major problems in modern concrete buildings, especially in locations that are difficult to map manually, such as bridges and high-rise buildings. Accurate analysis of unmanned aerial vehicle (UAV) images has become the key to determining whether a building needs maintenance. Methods:: Traditional image processing methods are easily interfered by high-frequency background. Neural network methods need fine datasets, which increase labor costs. Therefore, this paper proposes a segmentation algorithm based on UNet3+ network. After obtaining the UAV image, the rough location of the crack can be obtained by only rough labeling. And then, the sample balance can be carried out by clipping the target area. The UNet3+ network is used to train the processed datasets and extract the region of interest to ignore the non-target texture. Finally, the region of interest is further segmented by color clustering and edge detection methods. Results:: The proposed method can detect the cracks accurately. In all test images, the relative errors are less than 13%. Especially in test images whose width is less than 0.2mm, the maximum absolute error is only 0.0237mm, which is completely acceptable in actual production. The proposed method has higher practicability in the detection of concrete crack images taken by UAV. The results show that the proposed method outperforms the cutting-edge method published in the journal "Sensor", when the background is complex. Conclusion:: The proposed method can segment and detect cracks effectively, which can remove the high-frequency interference region from the images.

A method for cabbage root posture recognition based on YOLOv5s

Efficient, non-destructive cabbage harvesting is crucial for preserving its flavor and quality. Current cabbage harvesting mainly relies on mechanized automatic picking methods. However, a notable deficiency in most existing cabbage harvesting devices is the absence of a root posture recognition system to promptly adjust the root posture, consequently impacting the accuracy of root cutting during harvesting. To address this issue, this study introduces a cabbage root posture recognition method that combines deep learning with traditional image processing algorithms. Preliminary detection of the main root Region of Interest (ROI) areas of the cabbage is carried out through the YOLOv5s deep learning model. Subsequently, traditional image processing methods, the Graham algorithm, and the method of calculating the minimum circumscribed rectangle are employed to specifically detect the inclination angle of cabbage roots. This approach effectively addresses the difficulty in calculating the inclination angle of roots caused by occlusion from outer leaves. The results demonstrate that the precision and recall of this method are 98.7 % and 98.6 %, respectively, with an average absolute error of 0.80° and an average relative error of 1.34 % in posture. Using this method as a reference for mechanical harvesting can effectively mitigate cabbage damage rates.