The advent of artificial intelligence(AI)has propelled augmented reality(AR)display technology to a pivotal juncture,positioning it as a contender for the next generation of mobile intelligent terminals.However,the pu...The advent of artificial intelligence(AI)has propelled augmented reality(AR)display technology to a pivotal juncture,positioning it as a contender for the next generation of mobile intelligent terminals.However,the pursuit of advanced AR displays,particularly those capable of delivering immersive 3D experiences,is significantly hindered by the performance limitations of current hardware and the complexity of system integration.In this study,we present an innovative multi-focal plane AR display system that integrates a non-orthogonal polarization-multiplexing metasurface,freeform optical elements,and an OLED display screen.All optical elements are integrated into a single solid-state architecture,based on a joint optimization design approach of ray tracing and diffraction theory.The multi-focal plane AR visual effect is realized by the compact and multiplexing metasurface,which performs distinct phase functions across diverse polarization channels.Meanwhile,freeform surfaces offer ample design flexibility for the collaborative optimization of multi-focal plane imaging and the see-through systems.Followed by a mechanical design and prototype assembly,we demonstrate the system's capabilities in real-time and multi-focal plane display.The digital images at all virtual image distances seamlessly integrate with the real environment,fully exhibiting the system's high parallelism and real-time interactivity.With the innovative design concept and joint design method,we believe that our work will spur more innovative and compact intelligent solutions for AR displays and inject new vitality into hybrid optical systems.展开更多
Legal case classification involves the categorization of legal documents into predefined categories,which facilitates legal information retrieval and case management.However,real-world legal datasets often suffer from...Legal case classification involves the categorization of legal documents into predefined categories,which facilitates legal information retrieval and case management.However,real-world legal datasets often suffer from class imbalances due to the uneven distribution of case types across legal domains.This leads to biased model performance,in the form of high accuracy for overrepresented categories and underperformance for minority classes.To address this issue,in this study,we propose a data augmentation method that masks unimportant terms within a document selectively while preserving key terms fromthe perspective of the legal domain.This approach enhances data diversity and improves the generalization capability of conventional models.Our experiments demonstrate consistent improvements achieved by the proposed augmentation strategy in terms of accuracy and F1 score across all models,validating the effectiveness of the proposed method in legal case classification.展开更多
Surgical navigation has evolved significantly through advances in augmented reality,virtual reality,and mixed reality,improving precision and safety across many clinical applications,including neurosurgery,maxillofaci...Surgical navigation has evolved significantly through advances in augmented reality,virtual reality,and mixed reality,improving precision and safety across many clinical applications,including neurosurgery,maxillofacial,spinal,and arthroplasty procedures.By integrating preoperative imaging with real-time intraoperative data,these systems provide dynamic guidance,reduce radiation exposure,and minimize tissue damage.Key challenges persist,including intraoperative registration accuracy,flexible tissue deformation,respiratory compensation,and real-time imaging quality.Emerging solutions include artificial intelligence-driven segmentation,deformation-field modeling,and hybrid registration techniques.Future developments will include lightweight,portable systems,improved non-rigid registration algorithms,and greater clinical adoption.Despite advances in rigid-tissue applications,soft-tissue navigation requires additional innovation to address motion variability and registration reliability,ultimately advancing minimally invasive surgery and precision medicine.展开更多
Sign language is a primary mode of communication for individuals with hearing impairments,conveying meaning through hand shapes and hand movements.Contrary to spoken or written languages,sign language relies on the re...Sign language is a primary mode of communication for individuals with hearing impairments,conveying meaning through hand shapes and hand movements.Contrary to spoken or written languages,sign language relies on the recognition and interpretation of hand gestures captured in video data.However,sign language datasets remain relatively limited compared to those of other languages,which hinders the training and performance of deep learning models.Additionally,the distinct word order of sign language,unlike that of spoken language,requires context-aware and natural sentence generation.To address these challenges,this study applies data augmentation techniques to build a Korean Sign Language dataset and train recognition models.Recognized words are then reconstructed into complete sentences.The sign recognition process uses OpenCV and MediaPipe to extract hand landmarks from sign language videos and analyzes hand position,orientation,and motion.The extracted features are converted into time-series data and fed into a Long Short-Term Memory(LSTM)model.The proposed recognition framework achieved an accuracy of up to 81.25%,while the sentence generation achieved an accuracy of up to 95%.The proposed approach is expected to be applicable not only to Korean Sign Language but also to other low-resource sign languages for recognition and translation tasks.展开更多
Quality control plays a critical role in modern manufacturing.With the rapid development of electric vehicles,5G communications,and the semiconductor industry,high-speed and high-precision detection of surface defects...Quality control plays a critical role in modern manufacturing.With the rapid development of electric vehicles,5G communications,and the semiconductor industry,high-speed and high-precision detection of surface defects on silicon carbide(SiC)wafers has become essential.This study developed an automated inspection framework for identifying surface defects on SiC wafers during the coarse grinding stage.Thecomplex machining textures on wafer surfaces hinder conventional machine vision models,often leading to misjudgment.To address this,deep learning algorithms were applied for defect classification.Because defects are rare and imbalanced across categories,data augmentation was performed using aWasserstein generative adversarial network with gradient penalty(WGAN-GP),along with conventionalmethods.An improved YOLOv8-seg instance segmentationmodel was then trained and tested on datasets with different augmentation strategies.Experimental results showed that,when trained withWGAN-GP–generated data,YOLOv8-seg achieved mean average precision values of 87.0%(bounding box)and 86.6%(segmentation mask).Compared with the traditional WGAN-GP,the proposed model reduced Frechet inception distance by 32.2%and multiscale structural similarity index by 29.8%,generating more realistic and diverse defect images.The proposed framework effectively improves defect detection accuracy under limited data conditions and shows strong potential for industrial applications.展开更多
IoT devices are highly vulnerable to cyberattacks due to their widespread,distributed nature and limited security features.Intrusion detection can counter these threats,but class imbalance between normal and abnormal ...IoT devices are highly vulnerable to cyberattacks due to their widespread,distributed nature and limited security features.Intrusion detection can counter these threats,but class imbalance between normal and abnormal traffic often degrades model performance.We propose a novel multi-generator adversarial data augmentation method that blends the strengths of TMG-GAN(Tabular Multi-Generator Generative Adversarial Network)and R3GAN(Re-GAN).Our approach uses multiple class-specific generators to create diverse,high-quality synthetic samples,improving training stability and minority-class detection.A dual-branch discriminator-classifier enhances authenticity and class prediction,while feature similarity and decoupling techniques ensure clear class separation.Experiments on TON-IoT and Edge-IIoTset datasets show our method outperforms existing techniques like hybrid sampling,SNGAN(Spectral Normalization GAN),and TMG-GAN,achieving higher detection accuracy and better minority-class recall for imbalanced IoT intrusion detection.展开更多
To address the issue of inconsistent image quality and data scarcity in bolt defect detection for transmission lines,this paper proposes an improved sparse region-based convolutional neural network(RCNN) based detecti...To address the issue of inconsistent image quality and data scarcity in bolt defect detection for transmission lines,this paper proposes an improved sparse region-based convolutional neural network(RCNN) based detection framework integrating image quality evaluation and text-to-image data augmentation.First,a HyperNetwork-based image quality assessment module is introduced to filter low-quality inspection images in terms of clarity and structural integrity,resulting in a high-quality training dataset.Second,a text-to-image diffusion model is utilized for sample augmentation.By designing text prompts that describe various bolt defect types under diverse lighting and viewing conditions,the model automatically generates realistic synthetic samples.The generated images are further filtered using a combination of quality and perceptual similarity metrics to ensure consistency with the real data distribution.Building upon the sparse RCNN baseline,a dynamic label assignment mechanism and a random decision path detection head are incorporated to enhance bounding box matching and prediction accuracy.Experimental results demonstrate that the proposed method significantly improves detection accuracy(mAP@0.5) over the original sparse RCNN while maintaining low computational cost,enabling more efficient and intelligent inspection of transmission line components.展开更多
An improved cycle-consistent generative adversarial network(CycleGAN) method for defect data augmentation based on feature fusion and self attention residual module is proposed to address the insufficiency of defect s...An improved cycle-consistent generative adversarial network(CycleGAN) method for defect data augmentation based on feature fusion and self attention residual module is proposed to address the insufficiency of defect sample data for light guide plate(LGP) in production,as well as the problem of minor defects.Two optimizations are made to the generator of CycleGAN:fusion of low resolution features obtained from partial up-sampling and down-sampling with high-resolution features,combination of self attention mechanism with residual network structure to replace the original residual module.Qualitative and quantitative experiments were conducted to compare different data augmentation methods,and the results show that the defect images of the LGP generated by the improved network were more realistic,and the accuracy of the you only look once version 5(YOLOv5) detection network for the LGP was improved by 5.6%,proving the effectiveness and accuracy of the proposed method.展开更多
Photonic platforms are gradually emerging as a promising option to encounter the ever-growing demand for artificial intelligence,among which photonic time-delay reservoir computing(TDRC)is widely anticipated.While suc...Photonic platforms are gradually emerging as a promising option to encounter the ever-growing demand for artificial intelligence,among which photonic time-delay reservoir computing(TDRC)is widely anticipated.While such a computing paradigm can only employ a single photonic device as the nonlinear node for data processing,the performance highly relies on the fading memory provided by the delay feedback loop(FL),which sets a restriction on the extensibility of physical implementation,especially for highly integrated chips.Here,we present a simplified photonic scheme for more flexible parameter configurations leveraging the designed quasi-convolution coding(QC),which completely gets rid of the dependence on FL.Unlike delay-based TDRC,encoded data in QC-based RC(QRC)enables temporal feature extraction,facilitating augmented memory capabilities.Thus,our proposed QRC is enabled to deal with time-related tasks or sequential data without the implementation of FL.Furthermore,we can implement this hardware with a low-power,easily integrable vertical-cavity surface-emitting laser for high-performance parallel processing.We illustrate the concept validation through simulation and experimental comparison of QRC and TDRC,wherein the simpler-structured QRC outperforms across various benchmark tasks.Our results may underscore an auspicious solution for the hardware implementation of deep neural networks.展开更多
Peri-implant keratinized mucosa(PIKM)augmentation refers to surgical procedures aimed at increasing the width of PIKM.Consensus reports emphasize the necessity of maintaining a minimum width of PIKM to ensure long-ter...Peri-implant keratinized mucosa(PIKM)augmentation refers to surgical procedures aimed at increasing the width of PIKM.Consensus reports emphasize the necessity of maintaining a minimum width of PIKM to ensure long-term peri-implant health.Currently,several surgical techniques have been validated for their effectiveness in increasing PIKM.However,the selection and application of PIKM augmentation methods may present challenges for dental practitioners due to heterogeneity in surgical techniques,variations in clinical scenarios,and anatomical differences.Therefore,clear guidelines and considerations for PIKM augmentation are needed.This expert consensus focuses on the commonly employed surgical techniques for PIKM augmentation and the factors influencing their selection at second-stage surgery.It aims to establish a standardized framework for assessing,planning,and executing PIKM augmentation procedures,with the goal of offering evidence-based guidance to enhance the predictability and success of PIKM augmentation.展开更多
The integration of image analysis through deep learning(DL)into rock classification represents a significant leap forward in geological research.While traditional methods remain invaluable for their expertise and hist...The integration of image analysis through deep learning(DL)into rock classification represents a significant leap forward in geological research.While traditional methods remain invaluable for their expertise and historical context,DL offers a powerful complement by enhancing the speed,objectivity,and precision of the classification process.This research explores the significance of image data augmentation techniques in optimizing the performance of convolutional neural networks(CNNs)for geological image analysis,particularly in the classification of igneous,metamorphic,and sedimentary rock types from rock thin section(RTS)images.This study primarily focuses on classic image augmentation techniques and evaluates their impact on model accuracy and precision.Results demonstrate that augmentation techniques like Equalize significantly enhance the model's classification capabilities,achieving an F1-Score of 0.9869 for igneous rocks,0.9884 for metamorphic rocks,and 0.9929 for sedimentary rocks,representing improvements compared to the baseline original results.Moreover,the weighted average F1-Score across all classes and techniques is 0.9886,indicating an enhancement.Conversely,methods like Distort lead to decreased accuracy and F1-Score,with an F1-Score of 0.949 for igneous rocks,0.954 for metamorphic rocks,and 0.9416 for sedimentary rocks,exacerbating the performance compared to the baseline.The study underscores the practicality of image data augmentation in geological image classification and advocates for the adoption of DL methods in this domain for automation and improved results.The findings of this study can benefit various fields,including remote sensing,mineral exploration,and environmental monitoring,by enhancing the accuracy of geological image analysis both for scientific research and industrial applications.展开更多
Large-scale point cloud datasets form the basis for training various deep learning networks and achieving high-quality network processing tasks.Due to the diversity and robustness constraints of the data,data augmenta...Large-scale point cloud datasets form the basis for training various deep learning networks and achieving high-quality network processing tasks.Due to the diversity and robustness constraints of the data,data augmentation(DA)methods are utilised to expand dataset diversity and scale.However,due to the complex and distinct characteristics of LiDAR point cloud data from different platforms(such as missile-borne and vehicular LiDAR data),directly applying traditional 2D visual domain DA methods to 3D data can lead to networks trained using this approach not robustly achieving the corresponding tasks.To address this issue,the present study explores DA for missile-borne LiDAR point cloud using a Monte Carlo(MC)simulation method that closely resembles practical application.Firstly,the model of multi-sensor imaging system is established,taking into account the joint errors arising from the platform itself and the relative motion during the imaging process.A distortion simulation method based on MC simulation for augmenting missile-borne LiDAR point cloud data is proposed,underpinned by an analysis of combined errors between different modal sensors,achieving high-quality augmentation of point cloud data.The effectiveness of the proposed method in addressing imaging system errors and distortion simulation is validated using the imaging scene dataset constructed in this paper.Comparative experiments between the proposed point cloud DA algorithm and the current state-of-the-art algorithms in point cloud detection and single object tracking tasks demonstrate that the proposed method can improve the network performance obtained from unaugmented datasets by over 17.3%and 17.9%,surpassing SOTA performance of current point cloud DA algorithms.展开更多
Data augmentation plays an important role in training deep neural model by expanding the size and diversity of the dataset.Initially,data augmentation mainly involved some simple transformations of images.Later,in ord...Data augmentation plays an important role in training deep neural model by expanding the size and diversity of the dataset.Initially,data augmentation mainly involved some simple transformations of images.Later,in order to increase the diversity and complexity of data,more advanced methods appeared and evolved to sophisticated generative models.However,these methods required a mass of computation of training or searching.In this paper,a novel training-free method that utilises the Pre-Trained Segment Anything Model(SAM)model as a data augmentation tool(PTSAM-DA)is proposed to generate the augmented annotations for images.Without the need for training,it obtains prompt boxes from the original annotations and then feeds the boxes to the pre-trained SAM to generate diverse and improved annotations.In this way,annotations are augmented more ingenious than simple manipulations without incurring huge computation for training a data augmentation model.Multiple comparative experiments on three datasets are conducted,including an in-house dataset,ADE20K and COCO2017.On this in-house dataset,namely Agricultural Plot Segmentation Dataset,maximum improvements of 3.77%and 8.92%are gained in two mainstream metrics,mIoU and mAcc,respectively.Consequently,large vision models like SAM are proven to be promising not only in image segmentation but also in data augmentation.展开更多
In order to address the widespread data shortage problem in battery research,this paper proposes a generative adversarial network model that combines it with deep convolutional networks,the Wasserstein distance,and th...In order to address the widespread data shortage problem in battery research,this paper proposes a generative adversarial network model that combines it with deep convolutional networks,the Wasserstein distance,and the gradient penalty to achieve data augmentation.To lower the threshold for implementing the proposed method,transfer learning is further introduced.The W-DC-GAN-GP-TL framework is thereby formed.This framework is evaluated on 3 different publicly available datasets to judge the quality of generated data.Through visual comparisons and the examination of two visualization methods(probability density function(PDF)and principal component analysis(PCA)),it is demonstrated that the generated data is hard to distinguish from the real data.The application of generated data for training a battery state model using transfer learning is further evaluated.Specifically,Bi-GRU-based and Transformer-based methods are implemented on 2 separate datasets for estimating state of health(SOH)and state of charge(SOC),respectively.The results indicate that the proposed framework demonstrates satisfactory performance in different scenarios:for the data replacement scenario,where real data are removed and replaced with generated data,the state estimator accuracy decreases only slightly;for the data enhancement scenario,the estimator accuracy is further improved.The estimation accuracy of SOH and SOC is as low as 0.69%and 0.58%root mean square error(RMSE)after applying the proposed framework.This framework provides a reliable method for enriching battery measurement data.It is a generalized framework capable of generating a variety of time series data.展开更多
Existing semi-supervisedmedical image segmentation algorithms use copy-paste data augmentation to correct the labeled-unlabeled data distribution mismatch.However,current copy-paste methods have three limitations:(1)t...Existing semi-supervisedmedical image segmentation algorithms use copy-paste data augmentation to correct the labeled-unlabeled data distribution mismatch.However,current copy-paste methods have three limitations:(1)training the model solely with copy-paste mixed pictures from labeled and unlabeled input loses a lot of labeled information;(2)low-quality pseudo-labels can cause confirmation bias in pseudo-supervised learning on unlabeled data;(3)the segmentation performance in low-contrast and local regions is less than optimal.We design a Stochastic Augmentation-Based Dual-Teaching Auxiliary Training Strategy(SADT),which enhances feature diversity and learns high-quality features to overcome these problems.To be more precise,SADT trains the Student Network by using pseudo-label-based training from Teacher Network 1 and supervised learning with labeled data,which prevents the loss of rare labeled data.We introduce a bi-directional copy-pastemask with progressive high-entropy filtering to reduce data distribution disparities and mitigate confirmation bias in pseudo-supervision.For the mixed images,Deep-Shallow Spatial Contrastive Learning(DSSCL)is proposed in the feature spaces of Teacher Network 2 and the Student Network to improve the segmentation capabilities in low-contrast and local areas.In this procedure,the features retrieved by the Student Network are subjected to a random feature perturbation technique.On two openly available datasets,extensive trials show that our proposed SADT performs much better than the state-ofthe-art semi-supervised medical segmentation techniques.Using only 10%of the labeled data for training,SADT was able to acquire a Dice score of 90.10%on the ACDC(Automatic Cardiac Diagnosis Challenge)dataset.展开更多
Augmented reality(AR)is an emerging dynamic technology that effectively supports education across different levels.The increased use of mobile devices has an even greater impact.As the demand for AR applications in ed...Augmented reality(AR)is an emerging dynamic technology that effectively supports education across different levels.The increased use of mobile devices has an even greater impact.As the demand for AR applications in education continues to increase,educators actively seek innovative and immersive methods to engage students in learning.However,exploring these possibilities also entails identifying and overcoming existing barriers to optimal educational integration.Concurrently,this surge in demand has prompted the identification of specific barriers,one of which is three-dimensional(3D)modeling.Creating 3D objects for augmented reality education applications can be challenging and time-consuming for the educators.To address this,we have developed a pipeline that creates realistic 3D objects from the two-dimensional(2D)photograph.Applications for augmented and virtual reality can then utilize these created 3D objects.We evaluated the proposed pipeline based on the usability of the 3D object and performance metrics.Quantitatively,with 117 respondents,the co-creation team was surveyed with openended questions to evaluate the precision of the 3D object created by the proposed photogrammetry pipeline.We analyzed the survey data using descriptive-analytical methods and found that the proposed pipeline produces 3D models that are positively accurate when compared to real-world objects,with an average mean score above 8.This study adds new knowledge in creating 3D objects for augmented reality applications by using the photogrammetry technique;finally,it discusses potential problems and future research directions for 3D objects in the education sector.展开更多
In the context of power generation companies, vast amounts of specialized data and expert knowledge have been accumulated. However, challenges such as data silos and fragmented knowledge hinder the effective utilizati...In the context of power generation companies, vast amounts of specialized data and expert knowledge have been accumulated. However, challenges such as data silos and fragmented knowledge hinder the effective utilization of this information. This study proposes a novel framework for intelligent Question-and-Answer (Q&A) systems based on Retrieval-Augmented Generation (RAG) to address these issues. The system efficiently acquires domain-specific knowledge by leveraging external databases, including Relational Databases (RDBs) and graph databases, without additional fine-tuning for Large Language Models (LLMs). Crucially, the framework integrates a Dynamic Knowledge Base Updating Mechanism (DKBUM) and a Weighted Context-Aware Similarity (WCAS) method to enhance retrieval accuracy and mitigate inherent limitations of LLMs, such as hallucinations and lack of specialization. Additionally, the proposed DKBUM dynamically adjusts knowledge weights within the database, ensuring that the most recent and relevant information is utilized, while WCAS refines the alignment between queries and knowledge items by enhanced context understanding. Experimental validation demonstrates that the system can generate timely, accurate, and context-sensitive responses, making it a robust solution for managing complex business logic in specialized industries.展开更多
Cassava is the most widely distributed food crop in Central Africa. Chikwangue, also known as kwanga in the Republic of Congo, is a starchy fermented cassava product that is a staple food in the country. This work aim...Cassava is the most widely distributed food crop in Central Africa. Chikwangue, also known as kwanga in the Republic of Congo, is a starchy fermented cassava product that is a staple food in the country. This work aims to determine the composition of bioactive compounds in chikwangue, including biosurfactant-like molecules and proteins content. Antibacterial activities were investigated through the preliminary emulsification index of chikwangue and fermented paste. Antibacterial assay, 16S rRNA, cytK, hblD, nheB and entFM PCR amplifications, DNA sequence analysis, NCBI homology analysis, and phylogenic tree were performed using NGPhylogeny. fr and iTOL (interactive of live). Fermented cassava paste and chikwangue contain biosurfactants with an emulsification index of 50%. The total protein concentration in fermented cassava paste was 4 g/ml and the chikwangue was 2.5 g/mL Further sequence analysis showed that isolates shared a homology of up to 99.9% with Bacillus cereus PQ432941.1, B. licheniformis PQ432758.1, B. altitudinis PQ432754.1, B. subtilis PQ432759.1, B. mojavensis PQ432755.1, B. tequilensis MT994788.1, B. subtilis MT994789.1, Paenibacillus polymyxa PQ452544.1, B. velezensis PQ452545.1, B. thuringiensis PQ432763.1, B. pumilus PQ432762.1, B. subtilis MT994787.1, B. mycoides PQ432890.1, B. thuringiensis PQ432766.1, B. subtilis PQ432757.1 and B. amyloliquefaciens PQ432756.1. Importantly, the emulsification index (E24) ranged from 60 to 100% and the crude biosurfactant for the Bacillus strains mentioned above could easily inhibit the growth for pathogen Gram-negative bacteria (S. enterica, S. flexneri, E. coli, Klebsiella sp. and P. aeruginosa) with diameters ranging from 2.3 ± 0.1 cm to 5.5 ± 0.4 cm. On the other hand, the diameters of Gram-positive pathogenic bacteria (B. cereus and S. aureus) varied between 1.5 ± 0.5 cm and 4.0 ± 0.2 cm. These findings involve the promise purpose of Bacillus isolated from retted cassava, and this study systematically uncovered the biodiversity and distribution characteristics of retted paste cassava and chikwangue.展开更多
Rainfall-induced shallow landslides pose one of significant geological hazards,necessitating precise monitoring and prediction for effective disaster mitigation.Most studies on landslide prediction have focused on opt...Rainfall-induced shallow landslides pose one of significant geological hazards,necessitating precise monitoring and prediction for effective disaster mitigation.Most studies on landslide prediction have focused on optimizing machine learning(ML)algorithms,very limited attention has been paid to enhancing data quality for improved predictive performance.This study employs strategic data augmentation(DA)techniques to enhance the accuracy of shallow landslide prediction.Using five DA methods including singular spectrum analysis(SSA),moving averages(MA),wavelet denoising(WD),variational mode decomposition(VMD),and linear interpolation(LI),we utilize strategies such as smoothing,denoising,trend decomposition,and synthetic data generation to improve the training dataset.Four machine learning algorithms,i.e.artificial neural network(ANN),recurrent neural network(RNN),one-dimensional convolutional neural network(CNN1D),and long short-term memory(LSTM),are used to forecast landslide displacement.The case study of a landslide in southwest China shows the effectiveness of our approach in predicting landslide displacements,despite the inherent limitations of the monitoring dataset.VMD proves the most effective for smoothing and denoising,improving R^(2),RMSE,and MAPE by 172.16%,71.82%,and 98.9%,respectively.SSA addresses missing data,while LI is effective with limited data samples,improving metrics by 21.6%,52.59%,and 47.87%,respectively.This study demonstrates the potential of DA techniques to mitigate the impact of data defects on landslide prediction accuracy,with implications for similar cases.展开更多
Bird species classification is not only a challenging topic in artificial intelligence but also a domain closely related to environmental protection and ecological research.Additionally,performing edge computing on lo...Bird species classification is not only a challenging topic in artificial intelligence but also a domain closely related to environmental protection and ecological research.Additionally,performing edge computing on low-level devices using small neural networks can be an important research direction.In this paper,we use the EfficientNetV2B0 model for bird species classification,applying transfer learning on a dataset of 525 bird species.We also employ the BiRefNet model to remove backgrounds from images in the training set.The generated background-removed images are mixed with the original training set as a form of data augmentation.We aim for these background-removed images to help the model focus on key features,and by combining data augmentation with transfer learning,we trained a highly accurate and efficient bird species classification model.The training process is divided into a transfer learning stage and a fine-tuning stage.In the transfer learning stage,only the newly added custom layers are trained;while in the fine-tuning stage,all pre-trained layers except for the batch normalization layers are fine-tuned.According to the experimental results,the proposed model not only has an advantage in size compared to other models but also outperforms them in various metrics.The training results show that the proposed model achieved an accuracy of 99.54%and a precision of 99.62%,demonstrating that it achieves both lightweight design and high accuracy.To confirm the credibility of the results,we use heatmaps to interpret the model.The heatmaps show that our model can clearly highlight the image feature area.In addition,we also perform the 10-fold cross-validation on the model to verify its credibility.Finally,this paper proposes a model with low training cost and high accuracy,making it suitable for deployment on edge computing devices to provide lighter and more convenient services.展开更多
基金funding provided by National Natural Science Foundation of China(U21A20140)National Key Research and Development Program of China(2021YFA1401200)+2 种基金Beijing Natural Science Foundation(JQ24028)Beijing Nova Program(20240484557)Synergetic Extreme Condition User Facility(SECUF).
文摘The advent of artificial intelligence(AI)has propelled augmented reality(AR)display technology to a pivotal juncture,positioning it as a contender for the next generation of mobile intelligent terminals.However,the pursuit of advanced AR displays,particularly those capable of delivering immersive 3D experiences,is significantly hindered by the performance limitations of current hardware and the complexity of system integration.In this study,we present an innovative multi-focal plane AR display system that integrates a non-orthogonal polarization-multiplexing metasurface,freeform optical elements,and an OLED display screen.All optical elements are integrated into a single solid-state architecture,based on a joint optimization design approach of ray tracing and diffraction theory.The multi-focal plane AR visual effect is realized by the compact and multiplexing metasurface,which performs distinct phase functions across diverse polarization channels.Meanwhile,freeform surfaces offer ample design flexibility for the collaborative optimization of multi-focal plane imaging and the see-through systems.Followed by a mechanical design and prototype assembly,we demonstrate the system's capabilities in real-time and multi-focal plane display.The digital images at all virtual image distances seamlessly integrate with the real environment,fully exhibiting the system's high parallelism and real-time interactivity.With the innovative design concept and joint design method,we believe that our work will spur more innovative and compact intelligent solutions for AR displays and inject new vitality into hybrid optical systems.
基金supported by the Institute of Information&Communications Technology Planning&Evaluation(IITP)grant funded by the Korea government(MSIT)[RS-2021-II211341,Artificial Intelligence Graduate School Program(Chung-Ang University)],and by the Chung-Ang University Graduate Research Scholarship in 2024.
文摘Legal case classification involves the categorization of legal documents into predefined categories,which facilitates legal information retrieval and case management.However,real-world legal datasets often suffer from class imbalances due to the uneven distribution of case types across legal domains.This leads to biased model performance,in the form of high accuracy for overrepresented categories and underperformance for minority classes.To address this issue,in this study,we propose a data augmentation method that masks unimportant terms within a document selectively while preserving key terms fromthe perspective of the legal domain.This approach enhances data diversity and improves the generalization capability of conventional models.Our experiments demonstrate consistent improvements achieved by the proposed augmentation strategy in terms of accuracy and F1 score across all models,validating the effectiveness of the proposed method in legal case classification.
基金Supported by the National Natural Science Foundation of China(NSFC)under Grants 62025104,62422102,62331005,62301034,and U22A2052the Beijing Natural Science Foundation-Daxing Innovation Joint Fund(L256040).
文摘Surgical navigation has evolved significantly through advances in augmented reality,virtual reality,and mixed reality,improving precision and safety across many clinical applications,including neurosurgery,maxillofacial,spinal,and arthroplasty procedures.By integrating preoperative imaging with real-time intraoperative data,these systems provide dynamic guidance,reduce radiation exposure,and minimize tissue damage.Key challenges persist,including intraoperative registration accuracy,flexible tissue deformation,respiratory compensation,and real-time imaging quality.Emerging solutions include artificial intelligence-driven segmentation,deformation-field modeling,and hybrid registration techniques.Future developments will include lightweight,portable systems,improved non-rigid registration algorithms,and greater clinical adoption.Despite advances in rigid-tissue applications,soft-tissue navigation requires additional innovation to address motion variability and registration reliability,ultimately advancing minimally invasive surgery and precision medicine.
基金supported by the Institute of Information&Communications Technoljogy Planning&Evaluation(IITP)-Innovative Human Resource Development for Local Intellectualization Program grant funded by the Korea government(MSIT)(IITP-2026-RS-2022-00156334,50%)the National Research Foundation of Korea(NRF)grant funded by the Korea government(MSIT)(No.2021R1C1C2011105,50%).
文摘Sign language is a primary mode of communication for individuals with hearing impairments,conveying meaning through hand shapes and hand movements.Contrary to spoken or written languages,sign language relies on the recognition and interpretation of hand gestures captured in video data.However,sign language datasets remain relatively limited compared to those of other languages,which hinders the training and performance of deep learning models.Additionally,the distinct word order of sign language,unlike that of spoken language,requires context-aware and natural sentence generation.To address these challenges,this study applies data augmentation techniques to build a Korean Sign Language dataset and train recognition models.Recognized words are then reconstructed into complete sentences.The sign recognition process uses OpenCV and MediaPipe to extract hand landmarks from sign language videos and analyzes hand position,orientation,and motion.The extracted features are converted into time-series data and fed into a Long Short-Term Memory(LSTM)model.The proposed recognition framework achieved an accuracy of up to 81.25%,while the sentence generation achieved an accuracy of up to 95%.The proposed approach is expected to be applicable not only to Korean Sign Language but also to other low-resource sign languages for recognition and translation tasks.
基金funded by the National Science and Technology Council(NSTC),Taiwan,grant number NSTC 114-2218-E-167-001.
文摘Quality control plays a critical role in modern manufacturing.With the rapid development of electric vehicles,5G communications,and the semiconductor industry,high-speed and high-precision detection of surface defects on silicon carbide(SiC)wafers has become essential.This study developed an automated inspection framework for identifying surface defects on SiC wafers during the coarse grinding stage.Thecomplex machining textures on wafer surfaces hinder conventional machine vision models,often leading to misjudgment.To address this,deep learning algorithms were applied for defect classification.Because defects are rare and imbalanced across categories,data augmentation was performed using aWasserstein generative adversarial network with gradient penalty(WGAN-GP),along with conventionalmethods.An improved YOLOv8-seg instance segmentationmodel was then trained and tested on datasets with different augmentation strategies.Experimental results showed that,when trained withWGAN-GP–generated data,YOLOv8-seg achieved mean average precision values of 87.0%(bounding box)and 86.6%(segmentation mask).Compared with the traditional WGAN-GP,the proposed model reduced Frechet inception distance by 32.2%and multiscale structural similarity index by 29.8%,generating more realistic and diverse defect images.The proposed framework effectively improves defect detection accuracy under limited data conditions and shows strong potential for industrial applications.
基金Supported by the Key R&D Projects in Hubei Province(2025BAB018,2022BAA041)and Wuhan University Comprehensive Undergraduate Education Quality Reform Project。
文摘IoT devices are highly vulnerable to cyberattacks due to their widespread,distributed nature and limited security features.Intrusion detection can counter these threats,but class imbalance between normal and abnormal traffic often degrades model performance.We propose a novel multi-generator adversarial data augmentation method that blends the strengths of TMG-GAN(Tabular Multi-Generator Generative Adversarial Network)and R3GAN(Re-GAN).Our approach uses multiple class-specific generators to create diverse,high-quality synthetic samples,improving training stability and minority-class detection.A dual-branch discriminator-classifier enhances authenticity and class prediction,while feature similarity and decoupling techniques ensure clear class separation.Experiments on TON-IoT and Edge-IIoTset datasets show our method outperforms existing techniques like hybrid sampling,SNGAN(Spectral Normalization GAN),and TMG-GAN,achieving higher detection accuracy and better minority-class recall for imbalanced IoT intrusion detection.
基金Supported by the Science and Technology Project from State Grid Corporation of China (No.5700-202490330A-2-1-ZX)。
文摘To address the issue of inconsistent image quality and data scarcity in bolt defect detection for transmission lines,this paper proposes an improved sparse region-based convolutional neural network(RCNN) based detection framework integrating image quality evaluation and text-to-image data augmentation.First,a HyperNetwork-based image quality assessment module is introduced to filter low-quality inspection images in terms of clarity and structural integrity,resulting in a high-quality training dataset.Second,a text-to-image diffusion model is utilized for sample augmentation.By designing text prompts that describe various bolt defect types under diverse lighting and viewing conditions,the model automatically generates realistic synthetic samples.The generated images are further filtered using a combination of quality and perceptual similarity metrics to ensure consistency with the real data distribution.Building upon the sparse RCNN baseline,a dynamic label assignment mechanism and a random decision path detection head are incorporated to enhance bounding box matching and prediction accuracy.Experimental results demonstrate that the proposed method significantly improves detection accuracy(mAP@0.5) over the original sparse RCNN while maintaining low computational cost,enabling more efficient and intelligent inspection of transmission line components.
基金supported by the Jiangsu Province IUR Cooperation Project (No.BY2021258)the Wuxi Science and Technology Development Fund Project (No.G20212028)。
文摘An improved cycle-consistent generative adversarial network(CycleGAN) method for defect data augmentation based on feature fusion and self attention residual module is proposed to address the insufficiency of defect sample data for light guide plate(LGP) in production,as well as the problem of minor defects.Two optimizations are made to the generator of CycleGAN:fusion of low resolution features obtained from partial up-sampling and down-sampling with high-resolution features,combination of self attention mechanism with residual network structure to replace the original residual module.Qualitative and quantitative experiments were conducted to compare different data augmentation methods,and the results show that the defect images of the LGP generated by the improved network were more realistic,and the accuracy of the you only look once version 5(YOLOv5) detection network for the LGP was improved by 5.6%,proving the effectiveness and accuracy of the proposed method.
基金National Natural Science Foundation of China(62171305,62405206,62004135,62001317,62111530301)Natural Science Foundation of Jiangsu Province(BK20240778,BK20241917)+3 种基金State Key Laboratory of Advanced Optical Communication Systems and Networks,China(2023GZKF08)China Postdoctoral Science Foundation(2024M752314)Postdoctoral Fellowship Program of CPSF(GZC20231883)Innovative and Entrepreneurial Talent Program of Jiangsu Province(JSSCRC2021527).
文摘Photonic platforms are gradually emerging as a promising option to encounter the ever-growing demand for artificial intelligence,among which photonic time-delay reservoir computing(TDRC)is widely anticipated.While such a computing paradigm can only employ a single photonic device as the nonlinear node for data processing,the performance highly relies on the fading memory provided by the delay feedback loop(FL),which sets a restriction on the extensibility of physical implementation,especially for highly integrated chips.Here,we present a simplified photonic scheme for more flexible parameter configurations leveraging the designed quasi-convolution coding(QC),which completely gets rid of the dependence on FL.Unlike delay-based TDRC,encoded data in QC-based RC(QRC)enables temporal feature extraction,facilitating augmented memory capabilities.Thus,our proposed QRC is enabled to deal with time-related tasks or sequential data without the implementation of FL.Furthermore,we can implement this hardware with a low-power,easily integrable vertical-cavity surface-emitting laser for high-performance parallel processing.We illustrate the concept validation through simulation and experimental comparison of QRC and TDRC,wherein the simpler-structured QRC outperforms across various benchmark tasks.Our results may underscore an auspicious solution for the hardware implementation of deep neural networks.
基金supported by the Natural Science Foundation of Sichuan Province(grant number:25NSFSC0265).
文摘Peri-implant keratinized mucosa(PIKM)augmentation refers to surgical procedures aimed at increasing the width of PIKM.Consensus reports emphasize the necessity of maintaining a minimum width of PIKM to ensure long-term peri-implant health.Currently,several surgical techniques have been validated for their effectiveness in increasing PIKM.However,the selection and application of PIKM augmentation methods may present challenges for dental practitioners due to heterogeneity in surgical techniques,variations in clinical scenarios,and anatomical differences.Therefore,clear guidelines and considerations for PIKM augmentation are needed.This expert consensus focuses on the commonly employed surgical techniques for PIKM augmentation and the factors influencing their selection at second-stage surgery.It aims to establish a standardized framework for assessing,planning,and executing PIKM augmentation procedures,with the goal of offering evidence-based guidance to enhance the predictability and success of PIKM augmentation.
文摘The integration of image analysis through deep learning(DL)into rock classification represents a significant leap forward in geological research.While traditional methods remain invaluable for their expertise and historical context,DL offers a powerful complement by enhancing the speed,objectivity,and precision of the classification process.This research explores the significance of image data augmentation techniques in optimizing the performance of convolutional neural networks(CNNs)for geological image analysis,particularly in the classification of igneous,metamorphic,and sedimentary rock types from rock thin section(RTS)images.This study primarily focuses on classic image augmentation techniques and evaluates their impact on model accuracy and precision.Results demonstrate that augmentation techniques like Equalize significantly enhance the model's classification capabilities,achieving an F1-Score of 0.9869 for igneous rocks,0.9884 for metamorphic rocks,and 0.9929 for sedimentary rocks,representing improvements compared to the baseline original results.Moreover,the weighted average F1-Score across all classes and techniques is 0.9886,indicating an enhancement.Conversely,methods like Distort lead to decreased accuracy and F1-Score,with an F1-Score of 0.949 for igneous rocks,0.954 for metamorphic rocks,and 0.9416 for sedimentary rocks,exacerbating the performance compared to the baseline.The study underscores the practicality of image data augmentation in geological image classification and advocates for the adoption of DL methods in this domain for automation and improved results.The findings of this study can benefit various fields,including remote sensing,mineral exploration,and environmental monitoring,by enhancing the accuracy of geological image analysis both for scientific research and industrial applications.
基金Postgraduate Innovation Top notch Talent Training Project of Hunan Province,Grant/Award Number:CX20220045Scientific Research Project of National University of Defense Technology,Grant/Award Number:22-ZZCX-07+2 种基金New Era Education Quality Project of Anhui Province,Grant/Award Number:2023cxcysj194National Natural Science Foundation of China,Grant/Award Numbers:62201597,62205372,1210456foundation of Hefei Comprehensive National Science Center,Grant/Award Number:KY23C502。
文摘Large-scale point cloud datasets form the basis for training various deep learning networks and achieving high-quality network processing tasks.Due to the diversity and robustness constraints of the data,data augmentation(DA)methods are utilised to expand dataset diversity and scale.However,due to the complex and distinct characteristics of LiDAR point cloud data from different platforms(such as missile-borne and vehicular LiDAR data),directly applying traditional 2D visual domain DA methods to 3D data can lead to networks trained using this approach not robustly achieving the corresponding tasks.To address this issue,the present study explores DA for missile-borne LiDAR point cloud using a Monte Carlo(MC)simulation method that closely resembles practical application.Firstly,the model of multi-sensor imaging system is established,taking into account the joint errors arising from the platform itself and the relative motion during the imaging process.A distortion simulation method based on MC simulation for augmenting missile-borne LiDAR point cloud data is proposed,underpinned by an analysis of combined errors between different modal sensors,achieving high-quality augmentation of point cloud data.The effectiveness of the proposed method in addressing imaging system errors and distortion simulation is validated using the imaging scene dataset constructed in this paper.Comparative experiments between the proposed point cloud DA algorithm and the current state-of-the-art algorithms in point cloud detection and single object tracking tasks demonstrate that the proposed method can improve the network performance obtained from unaugmented datasets by over 17.3%and 17.9%,surpassing SOTA performance of current point cloud DA algorithms.
基金Natural Science Foundation of Zhejiang Province,Grant/Award Number:LY23F020025Science and Technology Commissioner Program of Huzhou,Grant/Award Number:2023GZ42Sichuan Provincial Science and Technology Support Program,Grant/Award Numbers:2023ZHCG0005,2023ZHCG0008。
文摘Data augmentation plays an important role in training deep neural model by expanding the size and diversity of the dataset.Initially,data augmentation mainly involved some simple transformations of images.Later,in order to increase the diversity and complexity of data,more advanced methods appeared and evolved to sophisticated generative models.However,these methods required a mass of computation of training or searching.In this paper,a novel training-free method that utilises the Pre-Trained Segment Anything Model(SAM)model as a data augmentation tool(PTSAM-DA)is proposed to generate the augmented annotations for images.Without the need for training,it obtains prompt boxes from the original annotations and then feeds the boxes to the pre-trained SAM to generate diverse and improved annotations.In this way,annotations are augmented more ingenious than simple manipulations without incurring huge computation for training a data augmentation model.Multiple comparative experiments on three datasets are conducted,including an in-house dataset,ADE20K and COCO2017.On this in-house dataset,namely Agricultural Plot Segmentation Dataset,maximum improvements of 3.77%and 8.92%are gained in two mainstream metrics,mIoU and mAcc,respectively.Consequently,large vision models like SAM are proven to be promising not only in image segmentation but also in data augmentation.
基金funded by the Bavarian State Ministry of Science,Research and Art(Grant number:H.2-F1116.WE/52/2)。
文摘In order to address the widespread data shortage problem in battery research,this paper proposes a generative adversarial network model that combines it with deep convolutional networks,the Wasserstein distance,and the gradient penalty to achieve data augmentation.To lower the threshold for implementing the proposed method,transfer learning is further introduced.The W-DC-GAN-GP-TL framework is thereby formed.This framework is evaluated on 3 different publicly available datasets to judge the quality of generated data.Through visual comparisons and the examination of two visualization methods(probability density function(PDF)and principal component analysis(PCA)),it is demonstrated that the generated data is hard to distinguish from the real data.The application of generated data for training a battery state model using transfer learning is further evaluated.Specifically,Bi-GRU-based and Transformer-based methods are implemented on 2 separate datasets for estimating state of health(SOH)and state of charge(SOC),respectively.The results indicate that the proposed framework demonstrates satisfactory performance in different scenarios:for the data replacement scenario,where real data are removed and replaced with generated data,the state estimator accuracy decreases only slightly;for the data enhancement scenario,the estimator accuracy is further improved.The estimation accuracy of SOH and SOC is as low as 0.69%and 0.58%root mean square error(RMSE)after applying the proposed framework.This framework provides a reliable method for enriching battery measurement data.It is a generalized framework capable of generating a variety of time series data.
基金supported by the Natural Science Foundation of China(No.41804112,author:Chengyun Song).
文摘Existing semi-supervisedmedical image segmentation algorithms use copy-paste data augmentation to correct the labeled-unlabeled data distribution mismatch.However,current copy-paste methods have three limitations:(1)training the model solely with copy-paste mixed pictures from labeled and unlabeled input loses a lot of labeled information;(2)low-quality pseudo-labels can cause confirmation bias in pseudo-supervised learning on unlabeled data;(3)the segmentation performance in low-contrast and local regions is less than optimal.We design a Stochastic Augmentation-Based Dual-Teaching Auxiliary Training Strategy(SADT),which enhances feature diversity and learns high-quality features to overcome these problems.To be more precise,SADT trains the Student Network by using pseudo-label-based training from Teacher Network 1 and supervised learning with labeled data,which prevents the loss of rare labeled data.We introduce a bi-directional copy-pastemask with progressive high-entropy filtering to reduce data distribution disparities and mitigate confirmation bias in pseudo-supervision.For the mixed images,Deep-Shallow Spatial Contrastive Learning(DSSCL)is proposed in the feature spaces of Teacher Network 2 and the Student Network to improve the segmentation capabilities in low-contrast and local areas.In this procedure,the features retrieved by the Student Network are subjected to a random feature perturbation technique.On two openly available datasets,extensive trials show that our proposed SADT performs much better than the state-ofthe-art semi-supervised medical segmentation techniques.Using only 10%of the labeled data for training,SADT was able to acquire a Dice score of 90.10%on the ACDC(Automatic Cardiac Diagnosis Challenge)dataset.
文摘Augmented reality(AR)is an emerging dynamic technology that effectively supports education across different levels.The increased use of mobile devices has an even greater impact.As the demand for AR applications in education continues to increase,educators actively seek innovative and immersive methods to engage students in learning.However,exploring these possibilities also entails identifying and overcoming existing barriers to optimal educational integration.Concurrently,this surge in demand has prompted the identification of specific barriers,one of which is three-dimensional(3D)modeling.Creating 3D objects for augmented reality education applications can be challenging and time-consuming for the educators.To address this,we have developed a pipeline that creates realistic 3D objects from the two-dimensional(2D)photograph.Applications for augmented and virtual reality can then utilize these created 3D objects.We evaluated the proposed pipeline based on the usability of the 3D object and performance metrics.Quantitatively,with 117 respondents,the co-creation team was surveyed with openended questions to evaluate the precision of the 3D object created by the proposed photogrammetry pipeline.We analyzed the survey data using descriptive-analytical methods and found that the proposed pipeline produces 3D models that are positively accurate when compared to real-world objects,with an average mean score above 8.This study adds new knowledge in creating 3D objects for augmented reality applications by using the photogrammetry technique;finally,it discusses potential problems and future research directions for 3D objects in the education sector.
文摘In the context of power generation companies, vast amounts of specialized data and expert knowledge have been accumulated. However, challenges such as data silos and fragmented knowledge hinder the effective utilization of this information. This study proposes a novel framework for intelligent Question-and-Answer (Q&A) systems based on Retrieval-Augmented Generation (RAG) to address these issues. The system efficiently acquires domain-specific knowledge by leveraging external databases, including Relational Databases (RDBs) and graph databases, without additional fine-tuning for Large Language Models (LLMs). Crucially, the framework integrates a Dynamic Knowledge Base Updating Mechanism (DKBUM) and a Weighted Context-Aware Similarity (WCAS) method to enhance retrieval accuracy and mitigate inherent limitations of LLMs, such as hallucinations and lack of specialization. Additionally, the proposed DKBUM dynamically adjusts knowledge weights within the database, ensuring that the most recent and relevant information is utilized, while WCAS refines the alignment between queries and knowledge items by enhanced context understanding. Experimental validation demonstrates that the system can generate timely, accurate, and context-sensitive responses, making it a robust solution for managing complex business logic in specialized industries.
文摘Cassava is the most widely distributed food crop in Central Africa. Chikwangue, also known as kwanga in the Republic of Congo, is a starchy fermented cassava product that is a staple food in the country. This work aims to determine the composition of bioactive compounds in chikwangue, including biosurfactant-like molecules and proteins content. Antibacterial activities were investigated through the preliminary emulsification index of chikwangue and fermented paste. Antibacterial assay, 16S rRNA, cytK, hblD, nheB and entFM PCR amplifications, DNA sequence analysis, NCBI homology analysis, and phylogenic tree were performed using NGPhylogeny. fr and iTOL (interactive of live). Fermented cassava paste and chikwangue contain biosurfactants with an emulsification index of 50%. The total protein concentration in fermented cassava paste was 4 g/ml and the chikwangue was 2.5 g/mL Further sequence analysis showed that isolates shared a homology of up to 99.9% with Bacillus cereus PQ432941.1, B. licheniformis PQ432758.1, B. altitudinis PQ432754.1, B. subtilis PQ432759.1, B. mojavensis PQ432755.1, B. tequilensis MT994788.1, B. subtilis MT994789.1, Paenibacillus polymyxa PQ452544.1, B. velezensis PQ452545.1, B. thuringiensis PQ432763.1, B. pumilus PQ432762.1, B. subtilis MT994787.1, B. mycoides PQ432890.1, B. thuringiensis PQ432766.1, B. subtilis PQ432757.1 and B. amyloliquefaciens PQ432756.1. Importantly, the emulsification index (E24) ranged from 60 to 100% and the crude biosurfactant for the Bacillus strains mentioned above could easily inhibit the growth for pathogen Gram-negative bacteria (S. enterica, S. flexneri, E. coli, Klebsiella sp. and P. aeruginosa) with diameters ranging from 2.3 ± 0.1 cm to 5.5 ± 0.4 cm. On the other hand, the diameters of Gram-positive pathogenic bacteria (B. cereus and S. aureus) varied between 1.5 ± 0.5 cm and 4.0 ± 0.2 cm. These findings involve the promise purpose of Bacillus isolated from retted cassava, and this study systematically uncovered the biodiversity and distribution characteristics of retted paste cassava and chikwangue.
基金supported by the National Natural Science Foundation of China(Grant No.42101089)Sichuan Science and Technology Program(2022YFS0586)the Open Fund of Key Laboratory of Mountain Hazards and Earth Surface Processes Chinese Academy of Sciences.
文摘Rainfall-induced shallow landslides pose one of significant geological hazards,necessitating precise monitoring and prediction for effective disaster mitigation.Most studies on landslide prediction have focused on optimizing machine learning(ML)algorithms,very limited attention has been paid to enhancing data quality for improved predictive performance.This study employs strategic data augmentation(DA)techniques to enhance the accuracy of shallow landslide prediction.Using five DA methods including singular spectrum analysis(SSA),moving averages(MA),wavelet denoising(WD),variational mode decomposition(VMD),and linear interpolation(LI),we utilize strategies such as smoothing,denoising,trend decomposition,and synthetic data generation to improve the training dataset.Four machine learning algorithms,i.e.artificial neural network(ANN),recurrent neural network(RNN),one-dimensional convolutional neural network(CNN1D),and long short-term memory(LSTM),are used to forecast landslide displacement.The case study of a landslide in southwest China shows the effectiveness of our approach in predicting landslide displacements,despite the inherent limitations of the monitoring dataset.VMD proves the most effective for smoothing and denoising,improving R^(2),RMSE,and MAPE by 172.16%,71.82%,and 98.9%,respectively.SSA addresses missing data,while LI is effective with limited data samples,improving metrics by 21.6%,52.59%,and 47.87%,respectively.This study demonstrates the potential of DA techniques to mitigate the impact of data defects on landslide prediction accuracy,with implications for similar cases.
文摘Bird species classification is not only a challenging topic in artificial intelligence but also a domain closely related to environmental protection and ecological research.Additionally,performing edge computing on low-level devices using small neural networks can be an important research direction.In this paper,we use the EfficientNetV2B0 model for bird species classification,applying transfer learning on a dataset of 525 bird species.We also employ the BiRefNet model to remove backgrounds from images in the training set.The generated background-removed images are mixed with the original training set as a form of data augmentation.We aim for these background-removed images to help the model focus on key features,and by combining data augmentation with transfer learning,we trained a highly accurate and efficient bird species classification model.The training process is divided into a transfer learning stage and a fine-tuning stage.In the transfer learning stage,only the newly added custom layers are trained;while in the fine-tuning stage,all pre-trained layers except for the batch normalization layers are fine-tuned.According to the experimental results,the proposed model not only has an advantage in size compared to other models but also outperforms them in various metrics.The training results show that the proposed model achieved an accuracy of 99.54%and a precision of 99.62%,demonstrating that it achieves both lightweight design and high accuracy.To confirm the credibility of the results,we use heatmaps to interpret the model.The heatmaps show that our model can clearly highlight the image feature area.In addition,we also perform the 10-fold cross-validation on the model to verify its credibility.Finally,this paper proposes a model with low training cost and high accuracy,making it suitable for deployment on edge computing devices to provide lighter and more convenient services.
