Imbalanced Distribution Research Articles

Posture Performance Analysis with Multiple Pressure Sensors

This paper aims to analyze in detail the postural performance of individuals using a mat equipped with multiple pressure sensors. By collecting foot pressure data, processing and analyzing these data with artificial intelligence algorithms, imbalances in the weight distribution and posture type of individuals will be detected. The study aims to provide a new method in the field of posture performance assessment and improvement. The use of multi-sensor data aims to provide more precise and reliable results than traditional analysis methods and to fill the gaps in the literature. The pressure sensors to be placed under the mat are optimized in a grid system and the data collected from this system is processed with the help of an electronic system. The electronic design allows analog data to be converted into digital format and transmitted to a workstation. The quality and accuracy of the data directly affects the success of artificial intelligence algorithms. Therefore, the data preprocessing step involves cleaning missing or erroneous data and combining information from different sensors using multi-sensor fusion techniques. Thus, the sensor placement and data processing processes are focused on improving the overall accuracy and portability performance of the system.

Analisis Distribusi Zakat di Indonesia: Porsi Penyaluran Zakat Kepada Asnaf Pada Tahun 2023

This study aims to analyze the distribution of zakat in Indonesia, focusing on the portion of zakat distribution to each asnaf in 2023. Zakat, as one of the pillars of Islamic economics, has an important role in reducing social inequality and improving people's welfare. The distribution of zakat in Indonesia is managed by various zakat institutions, one of which is Baitulmaal Munzalan Indonesia. This study uses a descriptive qualitative approach with content analysis to describe the distribution of zakat to five asnaf, namely fakir, poor, fi sabilillah, muallaf, and ibn sabil. The data used was obtained from 35 branches of Baitulmaal Munzalan throughout Indonesia in 2023. Data analysis was carried out qualitatively with the stages of data reduction, data presentation, and conclusion drawn. The results of the study show that the allocation of zakat is more given to the poor and Fi Sabilillah, while the Fakir, Muallaf, and Ibn Sabil get a lower allocation. This study also identifies various factors that affect the imbalance in zakat distribution, such as limited data collection, priority of zakat institutions, and socio-economic factors. Recommendations are given to increase the equitable distribution of zakat in the future

Assessing Service Imbalances as Contributing Factors to Mobility Issues in the Metropolitan District of Quito, Ecuador

This article analyzes the service distribution imbalance within the Metropolitan District of Quito (DMQ) and its impact on urban mobility, aiming to propose strategies for more equitable territorial planning. The data were gathered from sources such as the National Institute of Statistics and Census (INEC), the Ministry of Health, the Ministry of Education, and OpenStreetMap. These data were integrated with GIS tools to model patterns of accessibility and mobility. Through a comprehensive approach, the study assessed education, banking services, employment, and healthcare, identifying how inequitable access to these services drives increased travel demand, especially in rural and peri-urban areas. In the education field, over 500 neighborhoods faced a shortage of institutions, compelling students to commute to other neighborhoods. For financial services, only 67% of neighborhoods had adequate access, with disparities across different socioeconomic zones. Additionally, employment-related mobility posed another challenge, with 88% of workers commuting outside their residential parish. Finally, access to healthcare was also unequal across the DMQ, particularly in peripheral areas where residents must travel long distances. In this context, it can be concluded that more efficient urban planning in the Metropolitan District of Quito (DMQ) is crucial to address imbalances in the distribution of services and enhance quality of life. Proposed strategies include establishing a land reserve, decentralizing services to underserved areas, integrating smart technologies, and promoting incentives for remote work, sustainable mobility, and public transport. These actions aim to foster greater territorial equity and accessibility.

Additives Capable of Stably Supplying Anions/Cations for Homogeneous Lithium Deposition/Stripping.

One of the important factors leading to lithium dendrites is a slow lithium-ion mass transport and imbalanced distribution of the Li+ concentration and nuclei sites on the anode surface. To achieve uniform lithium deposition during the charge and discharge process, we introduce a homemade new copolymer (with the quaternary ammonium group N3R+I- on its side chain as the main functional group), named P35, as a functional electrolyte additive to regulate the lithium deposition. Theoretical calculations show that under the strong coordinating interaction between I- and N3R+, P35 preferentially adsorbs onto the lithium foil surface, effectively countering the adsorption of lithium salt anions such as PF6-. Moreover, the positive charge carried by the quaternary ammonium salt group of P35 could interact with PF6- to limit their mobility. Consequently, the dipole interaction on lithium ions is diminished, leading to an enhancement in the transport rate and a decrease in the concentration gradient of lithium ions. Furthermore, a more efficient SEI was formed due to the dual charges electrostatic shield formed by N3R+I-. Li-Li symmetric cells and Li-LiFePO4 full cells assembled with electrolytes with P35 exhibit better rate performance, smaller polarization, and smoother deposition morphology in comparison to the cells without the P35 additive.

Investigating How Lysophosphatidylcholine and Lysophosphatidylethanolamine Enhance the Membrane Permeabilization Efficacy of Host Defense Peptide Piscidin 1.

Lysophospholipids (LPLs) and host defense peptides (HDPs) are naturally occurring membrane-active agents that disrupt key membrane properties, including the hydrocarbon thickness, intrinsic curvature, and molecular packing. Although the membrane activity of these agents has been widely examined separately, their combined effects are largely unexplored. Here, we use experimental and computational tools to investigate how lysophosphatidylcholine (LPC) and lysophosphatidylethanolamine (LPE), an LPL of lower positive spontaneous curvature, influence the membrane activity of piscidin 1 (P1), an α-helical HDP from fish. Four membrane systems are probed: 75:25 C16:0-C18:1 PC (POPC)/C16:0-C18:1 phosphoglycerol (POPG), 50:25:25 POPC/POPG/16:0 LPC, 75:25 C16:0-C18:1 PE (POPE)/POPG, and 50:25:25 POPE/POPG/14:0 LPE. Dye leakage, circular dichroism, and NMR experiments demonstrate that while the presence of LPLs alone does not induce leakage-proficient defects, it boosts the permeabilization capability of P1, resulting in an efficacy order of POPC/POPG/16:0 LPC > POPE/POPG/14:0 LPE > POPC/POPG > POPE/POPG. This enhancement occurs without altering the membrane affinity and conformation of P1. Molecular dynamics simulations feature two types of asymmetric membranes to represent the imbalanced ("area stressed") and balanced ("area relaxed") distribution of lipids and peptides in the two leaflets. The simulations capture the membrane thinning effects of P1, LPC, and LPE, and the positive curvature strain imposed by both LPLs is reflected in the lateral pressure profiles. They also reveal a higher number of membrane defects for the P1/LPC than P1/LPE combination, congruent with the permeabilization experiments. Altogether, these results show that P1 and LPLs disrupt membranes in a concerted fashion, with LPC, the more disruptive LPL, boosting the permeabilization of P1 more than LPE. This mechanistic knowledge is relevant to understanding biological processes where multiple membrane-active agents such as HDPs and LPLs are involved.

Cloud-Edge-End Collaborative Federated Learning: Enhancing Model Accuracy and Privacy in Non-IID Environments.

Cloud-edge-end computing architecture is crucial for large-scale edge data processing and analysis. However, the diversity of terminal nodes and task complexity in this architecture often result in non-independent and identically distributed (non-IID) data, making it challenging to balance data heterogeneity and privacy protection. To address this, we propose a privacy-preserving federated learning method based on cloud-edge-end collaboration. Our method fully considers the three-tier architecture of cloud-edge-end systems and the non-IID nature of terminal node data. It enhances model accuracy while protecting the privacy of terminal node data. The proposed method groups terminal nodes based on the similarity of their data distributions and constructs edge subnetworks for training in collaboration with edge nodes, thereby mitigating the negative impact of non-IID data. Furthermore, we enhance WGAN-GP with attention mechanism to generate balanced synthetic data while preserving key patterns from original datasets, reducing the adverse effects of non-IID data on global model accuracy while preserving data privacy. In addition, we introduce data resampling and loss function weighting strategies to mitigate model bias caused by imbalanced data distribution. Experimental results on real-world datasets demonstrate that our proposed method significantly outperforms existing approaches in terms of model accuracy, F1-score, and other metrics.

Evaluating Inequity in Access to Park-Based Physical Activity at the Sub-District Scale: A Case Study in Xianyang, China

Urban densification has heightened residents’ demand for equitable access to urban park services, particularly those that support physical activity (PA). This access is crucial for public health and sustainable urban development. We developed the Park-Based Physical Activity Composite Attractiveness Score (PCAS) to assess the supply and service equity of parks at the sub-district scale in the central city of Xianyang, western China. The average PCAS for all parks in the study area was 46, with 69% scoring below the benchmark of 60, indicating that the parks fail to meet residents’ PA needs. We found a significant imbalance in park distribution. The 14 sub-districts we measured had an average supply score of 42 and a demand score of 49, demonstrating a clustering of park services in areas with natural resources and higher socioeconomic status. While the overall park service level was found to be reasonably equitable (Gini coefficient of 0.38), a large gap in the sub-dimensional indicators point to inequities in park services. The following scenario optimization suggested that adding new parks could more effectively enhance equity and residents’ PA levels than simply increasing park facilities. Our results provide valuable insights for urban park planning and policy-making, and contribute to the development of more equitable and accessible park services to achieve public health objectives.

Multi-expert contrastive learning for remote sensing long-tailed image classification

ABSTRACT Most existing remote sensing scene classification methods assume that the training data follows a balanced distribution. However, remote sensing data in the real world often exhibits imbalanced long-tailed distributions. The traditional deep learning model fits the head class well, but the tail class is not well fitted. In the face of complex background interference and the variability of similar land features, this manuscript proposes a multi-expert model based on contrastive learning and optimal transport to address these challenges. This method comprehensively considers the high inter-class similarity and large intra-class differences of long-tailed remote sensing data. By capturing key spectral discriminative features of land features in remote sensing scenes, the model’s generalization ability is enhanced to improve the performance of tail classes. Experiments with different balance ratios are conducted on three public datasets. They have verified that the proposed method achieves superior performance in remote sensing long-tailed scene classification tasks.

The mental health care system for children and adolescents in Greece: a review and structure assessment.

The mental health system in Greece faces challenges to complete its transition to a community-oriented model, having significant concerns for child and adolescent care due to lower coverage and service gaps. This component of the mental health system has not been comprehensively evaluated. We conducted a review of the mental health care system for children and adolescents in Greece. For a field assessment, we directly collected data from mental health services to map availability and distribution. We analyzed the needs of human resources using professional register data and the national census. The National Health Care Service (ESY, Εθνικό Σύστημα Υγείας) is the public health system in Greece, characterized by public governance but significant private participation. Although ESY aims for universal care, gaps in population coverage and high user fees create barriers to access. Embedded within ESY, the mental health system is shifting towards a community-oriented structure since the psychiatric reform. For children and adolescents, there is a developing framework for regionalization and community services, including day centers, inpatient facilities, outpatient departments, and school-based psychoeducational facilities. However, services lack coordination in a stepped care model. Patient pathways are not established and primary care rarely involves child mental health, leading to direct access to specialists. Services operate in isolation due to the absence of online registers. There is no systematic performance monitoring, yet some assessments indicate that professional practices may lack evidence-based guidelines. Our mapping highlighted a scarcity of public structures, with an unbalanced regional distribution and many underserved areas. Child and adolescent psychiatrists are predominantly affiliated with the private sector, leading to professional gaps in the public system. Our assessment identifies an established framework for a community-oriented, universally accessible mental health system, yet several barriers impede its full realization. These include an inconsistent primary healthcare system, a shortage of specialists in the public sector, imbalanced service distribution, lack of coordination among providers, underfunding, and absence of quality monitoring. We propose interventions to promote child and adolescent mental health in primary care, coordinate patient pathways, establish standards of care, and monitor performance.

Enhanced Intrusion Detection for ICS Using MS1DCNN and Transformer to Tackle Data Imbalance.

With the escalating threat posed by network intrusions, the development of efficient intrusion detection systems (IDSs) has become imperative. This study focuses on improving detection performance in programmable logic controller (PLC) network security while addressing challenges related to data imbalance and long-tail distributions. A dataset containing five types of attacks targeting programmable logic controllers (PLCs) in industrial control systems (ICS) was first constructed. To address class imbalance and challenges posed by complex network traffic, Synthetic Minority Oversampling Technique (SMOTE) and Borderline-SMOTE were applied to oversample minority classes, thereby enhancing their diversity. This paper proposes a dual-channel feature extraction model that integrates a multi-scale one-dimensional convolutional neural network (MS1DCNN) and a Weight-Dropped Transformer (WDTransformer) for IDS. The MS1DCNN is designed to extract fine-grained temporal features from packet-level data, whereas the WDTransformer leverages self-attention mechanisms to capture long-range dependencies and incorporates regularization techniques to mitigate overfitting. To further enhance performance on long-tail distributions, a custom combined loss function was developed by integrating cross-entropy loss and focal loss to reduce misclassification in minority classes. Experimental validation on the constructed dataset demonstrated that the proposed model achieved an accuracy of 95.11% and an F1 score of 95.12%, significantly outperforming traditional machine learning and deep learning models.

Res-LK-SLR: A Residual Network Based on Large Kernels and Shapelet-Level Representations for Spatial Infrared Spot Target Discrimination

Spatial infrared spot target (SIST) discrimination based on infrared radiation sequences (IRSs) can be considered a univariate trending time series classification task. However, due to the complexity of actual scenarios and the limited opportunities for acquiring IRSs, resulting in noise interference, extremely small-scale datasets with imbalanced distribution of classes and widely varying sequence lengths range from a few hundred to several thousand time steps. Current research is primarily based on idealized simulation datasets, resulting in a performance gap when applied to actual applications. To address these issues, firstly, we construct a simulation dataset tailored to the challenges of actual scenarios. Secondly, we design a practical data preprocessing method to achieve uniform sequence length, coarse alignment of shapelets and filtering while preserving key points. Thirdly, we propose a residual network Res-LK-SLR for IRS classification based on large kernels (LKs, providing long-term dependence) and shapelet-level representations (SLRs, where the hidden layer features are aligned with the learned high-level representations to obtain the optimal segmentation and generate shapelet-level representations). Additionally, we conduct extensive evaluations and validations on both the simulation dataset and 18 UCR time series classification datasets. The results demonstrate the effectiveness and generalization ability of our proposed Res-LK-SLR.

Coordination strategies to improve COVID-19 PCR laboratory testing scale up in Nepal: An analysis.

During the COVID-19 pandemic, Nepal rapidly expanded its PCR testing capacity, essential for effective outbreak response. However, many laboratories faced overwhelming test volumes, resulting in delays that may have hindered containment efforts. This study aims to determine whether these challenges stemmed from capacity limitations and/or imbalanced sample distribution due to inadequate coordination. In this retro-perspective simulation of SARS-CoV-2 testing in Nepal during 2021, we evaluate the impact of coordinated sample transfers on reducing laboratory stress and wait times during demand peaks. Our findings reveal that centralized coordination and strategic partnerships for sample transfers significantly enhance diagnostic network performance, even under high demand. These insights offer valuable guidance for policymakers on implementing effective coordination strategies to strengthen diagnostic networks for future pandemics.

Long-Term or Short-Term? Prediction of Ship Detention Duration Based on Machine Learning

The prevalence of ship deficiencies continues to be a significant issue. Data from the Tokyo Memorandum of Understanding reveals that ship detentions in 2023 surged by more than 80% compared with the previous year. The significant number of detained ships not only disrupts ships’ daily operations but also strains port resources, leading to increased additional costs. In light of this issue, predicting the duration of ship detention becomes crucial, as accurate predictions can assist port managers in resource allocation and provide shipping companies with critical information for operational planning. This study is the first to predict ship detention duration, specifically distinguishing between long-term and short-term detained ships. Initially, key deficiency types influencing the ship detention duration were identified using an improved entropy weight–grey relational analysis. Subsequently, in consideration of the imbalance in data distribution between long-term and short-term detentions, a random forest model capable of handling imbalanced data was applied to classify these two types. The study found that fire safety, propulsion and auxiliary machinery, and pollution prevention are the three most critical deficiency types impacting detention duration; and the random forest model sampled and processed from the data level possessed the best model performance, achieving prediction accuracies of 0.71, 0.72, and 0.85 for bulk carriers, containers, and oil tankers, respectively. This research offers a comprehensive analysis of ship detention duration, making a significant contribution to both the theoretical understanding and practical applications in the maritime industry. Accurately predicting ship detention duration provides valuable insights for stakeholders, enabling them to anticipate potential detention scenarios and thus supporting shipping companies in effective fleet management while assisting port authorities in the optimal allocation of berth resources.

The CWA Benchmark: A Seismic Dataset from Taiwan for Seismic Research

Abstract In recent years, numerous studies have employed deep learning in seismology, with data-driven neural network models increasingly becoming the norm in emerging research paradigms. Alongside advancements in model architectures, several benchmark datasets have also been introduced. However, many of these datasets suffer from imbalanced distributions, particularly a scarcity of large-magnitude events, which can impair the performance of deep learning-based models. Taiwan, situated in a seismically active region experiences a high frequency of earthquakes annually. The Central Weather Administration (CWA) in Taiwan maintains comprehensive records of these seismic events. This study introduces a benchmark dataset, named CWA, specifically compiled from this extensive database. The CWA benchmark includes 331 events with magnitudes greater than five, collected from a high-density seismic network spanning from 2011 to 2021, making it well-suited for training deep learning models. In addition, the CWA benchmark features over 40 attributes and ∼500,000 seismograms, providing valuable data labels for various seismology-related tasks.

Enhancing model robustness to imbalanced species abundance distributions: Eliminating misclassified records via a model-agnostic approach, exemplified by tuna fisheries datasets

Anomalies in species abundance data can potentially cause classification errors in ecological forecasting models. Accurate estimation of anomalies locations can enhance the predictive capacity of models. This study aims to propose an approach for precisely identifying and correcting anomalies within imbalanced species abundance data, thereby addressing the challenges posed by both anomalous and imbalanced species abundance distributions (SADs). A model-agnostic statistical tool, Confident Learning (CL) theory, is introduced to estimate the probability of each sample being misclassified during the prediction phase. Specifically, the approach targets classification errors from models trained on imbalanced SADs for data-cleansing, identifying these records as anomalies. The approach is applied to tuna fisheries datasets, focusing specifically on bigeye tuna (Thunnus obesus), a targeted species in longline fishing, and albacore tuna (Thunnus alalunga), a non-target species in the tropical Atlantic Ocean. These datasets, spanning from 2016 to 2019, featured a spatial resolution of 0.5° × 0.5° and daily temporal resolution, providing a comprehensive view of imbalanced data scenarios. The results demonstrate that all the predictors: Support Vector Machine (SVM), Logistic Regression (LR), Random Forests (RF), and Extreme Gradient Boosting (XGBoost) were considerably enhanced after training on the cleaned datasets. Notably, SVM and LR achieved overall accuracy rates of over 90% on predicting both low- and high-abundant fishing grounds. The proposed approach reveals that the elimination of anomalies can enhance the robustness of ecological forecasting models to imbalanced SADs, offering new insights and technical support for the delicate prediction and assessment of ecological resources.

Large-Scale Object Detection in the Wild With Imbalanced Data Distribution, and Multi-Labels.

Training with more data has always been the most stable and effective way of improving performance in the deep learning era. The Open Images dataset, the largest object detection dataset, presents significant opportunities and challenges for general and sophisticated scenarios. However, its semi-automatic collection and labeling process, designed to manage the huge data scale, leads to label-related problems, including explicit or implicit multiple labels per object and highly imbalanced label distribution. In this work, we quantitatively analyze the major problems in large-scale object detection and provide a detailed yet comprehensive demonstration of our solutions. First, we design a concurrent softmax to handle the multi-label problems in object detection and propose a soft-balance sampling method with a hybrid training scheduler to address the label imbalance. This approach yields a notable improvement of 3.34 points, achieving the best single-model performance with a mAP of 60.90% on the public object detection test set of Open Images. Then, we introduce a well-designed ensemble mechanism that substantially enhances the performance of the single model, achieving an overall mAP of 67.17%, which is 4.29 points higher than the best result from the Open Images public test 2018.

Complex product quality prediction based on ResNeSt model combining class-aware attention mechanism and dynamically updating class weights

Abstract The processing state data of complex products often exhibits high dimensionality, intricate feature relationships, and imbalanced distributions, hindering the accuracy of data-driven quality prediction models. To address these challenges, this paper proposes a complex product quality prediction model that integrates a class-aware attention mechanism and dynamic class weight update strategy. To handle the high dimensionality and complex correlations of the data, Pearson Correlation Coefficient and Gradient Boosting Decision Tree-based Recursive Feature Elimination (GBDT-RFE) algorithms are employed for feature selection. In response to the imbalanced distribution of data categories, a class-aware attention mechanism module is built upon the ResNeSt network, utilizing prior knowledge of data distribution to adjust the model’s focus. Furthermore, a loss function is designed to dynamically update class weights based on classification error rates, enabling the model to adaptively adjust the weight allocation for different classes, thereby enhancing its generalization capability. Experimental results on a semiconductor industry dataset demonstrate that the proposed model outperforms the original ResNeSt model in multiple metrics, with accuracy improved by 3.5%, AUC increased by 12.6%, F1 score raised by 18.9%, and recall enhanced by 24%, ultimately achieving an overall prediction accuracy of 98.7%. These multidimensional improvements make the model suitable for various complex product data scenarios such as electronics manufacturing, automotive industry, and biopharmaceuticals, demonstrating broad applicability.

Triplet Adaptation Framework for Robust Semi-Supervised Learning.

Semi-supervised learning (SSL) suffers from severe performance degradation when labeled and unlabeled data come from inconsistent and imbalanced distribution. Nonetheless, there is a lack of theoretical guidance regarding a remedy for this issue. To bridge the gap between theoretical insights and practical solutions, we embark to an analysis of generalization bound of classic SSL algorithms. This analysis reveals that distribution inconsistency between unlabeled and labeled data can cause a significant generalization error bound. Motivated by this theoretical insight, we present a Triplet Adaptation Framework (TAF) to reduce the distribution divergence and improve the generalization of SSL models. TAF comprises three adapters: Balanced Residual Adapter, aiming to map the class distribution of labeled and unlabeled data to a uniform distribution for reducing class distribution divergence; Representation Adapter, aiming to map the representation distribution of unlabeled data to labeled one for reducing representation distribution divergence; and Pseudo-Label Adapter, aiming to align the predicted pseudo-labels with the class distribution of unlabeled data, thereby preventing erroneous pseudo-labels from exacerbating representation divergence. These three adapters collaborate synergistically to reduce the generalization bound, ultimately achieving a more robust and generalizable SSL model. Extensive experiments across various robust SSL scenarios validate the efficacy of our method.

Instance Consistency Regularization for Semi-Supervised 3D Instance Segmentation.

Large-scale datasets with point-wise semantic and instance labels are crucial to 3D instance segmentation but also expensive. To leverage unlabeled data, previous semi-supervised 3D instance segmentation approaches have explored self-training frameworks, which rely on high-quality pseudo labels for consistency regularization. They intuitively utilize both instance and semantic pseudo labels in a joint learning manner. However, semantic pseudo labels contain numerous noise derived from the imbalanced category distribution and natural confusion of similar but distinct categories, which leads to severe collapses in self-training. Motivated by the observation that 3D instances are non-overlapping and spatially separable, we ask whether we can solely rely on instance consistency regularization for improved semi-supervised segmentation. To this end, we propose a novel self-training network InsTeacher3D to explore and exploit pure instance knowledge from unlabeled data. We first build a parallel base 3D instance segmentation model DKNet, which distinguishes each instance from the others via discriminative instance kernels without reliance on semantic segmentation. Based on DKNet, we further design a novel instance consistency regularization framework to generate and leverage high-quality instance pseudo labels. Experimental results on multiple large-scale datasets show that the InsTeacher3D significantly outperforms prior state-of-the-art semi-supervised approaches.

Teleseismic Full‐Waveform Tomography Based on a 3D FK‐LTSOS Hybrid Method for Local Imaging

AbstractTeleseismic waveforms contain abundant interpretable information about Earth's properties. They can be used to explore the refined structure of Earth's interior, especially in the regions with imbalanced spatial distribution of seismic activity. However, it's technically infeasible to numerically simulate high‐frequency (>1 Hz) teleseismic wave propagation within a whole domain iteratively in full‐waveform inversion due to its vast computational costs. We develop a 3D FK‐LTSOS (Frequency‐Wavenumber, Layered Time‐Space Optimized Symplectic) hybrid method and then apply it to teleseismic full‐waveform tomography to tackle this computational challenge. The 3D FK‐LTSOS hybrid method combines the semi‐analytical solution computed by the FK method rapidly in a 1D background model and the numerical solution calculated by the 3D LTSOS method accurately in 3D localized heterogeneous media with topography to simulate teleseismic wave propagation efficiently and accurately. The comparison of synthetic seismograms shows its accuracy and stability when simulating wave propagation in the topographic model. Based on this hybrid method, the teleseismic full‐waveform tomographic method is developed to efficiently resolve the detailed structure of the local research domain utilizing high‐frequency teleseismic data. The essential contents of full‐waveform tomography are presented, including misfit function, Fréchet kernels, smoothing strategy, and nonlinear conjugate gradient method. Synthetic data application for spherical anomaly models with planer surface and Gaussian topography, and observed data application for the crust‐upper mantle structure beneath eastern Tibet confirm the validity of the teleseismic full‐waveform tomography and demonstrate that our tomographic method can image the localized structures speedily from full‐waveform information.