Journal of Applied Geophysics最新文献

{"title":"Comprehensive detection method and assessment of karst development characteristics of diversion aqueduct: A Case Study in Yunnan, China","authors":"Zhi-Qiang Li ,&nbsp;Haoxuan Yan ,&nbsp;Lichao Nie ,&nbsp;Andong Chen ,&nbsp;Duo Zhang ,&nbsp;Xingbang Sun","doi":"10.1016/j.jappgeo.2025.105888","DOIUrl":"10.1016/j.jappgeo.2025.105888","url":null,"abstract":"<div><div>Due to the complex karst geological structure, the construction of diversion aqueducts in karst formations poses deformation and stability challenges for the foundation. During the construction of the aqueduct in Jifu Village, Yunnan Province, a comprehensive detection method was employed to investigate the karst development in the area. This method integrates information obtained from various techniques, including geological analysis, Ground Penetrating Radar (GPR), and drilling. Geological analysis was used to identify potential karst development locations within the project area. GPR provided geophysical profiles revealing shallow karst features, and drilling was strategically placed along these profiles to verify and supplement the GPR data. This approach yielded more accurate information on karst development. The results demonstrate that the combination of geological analysis, geophysics, and drilling can effectively detect karst formations. The identified karst locations were consistent with the inferred underground karst zones from the investigation. Optimizing and refining this method can offer valuable reference and guidance for detecting karst areas in similar regions.</div></div>","PeriodicalId":54882,"journal":{"name":"Journal of Applied Geophysics","volume":"242 ","pages":"Article 105888"},"PeriodicalIF":2.2,"publicationDate":"2025-07-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144713679","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Electrical resistivity monitoring of CO2 injection at the Mont Terri underground laboratory, Switzerland","authors":"Andrea Balza-Morales ,&nbsp;Melchior Grab ,&nbsp;Antonio Pio Rinaldi ,&nbsp;Alba Zappone ,&nbsp;Hansruedi Maurer ,&nbsp;Florian M. Wagner","doi":"10.1016/j.jappgeo.2025.105852","DOIUrl":"10.1016/j.jappgeo.2025.105852","url":null,"abstract":"<div><div>Monitoring CO<span><math><msub><mrow></mrow><mrow><mn>2</mn></mrow></msub></math></span> injection in the subsurface using geophysical methods, particularly in caprocks or hard rock, presents unique challenges. These challenges arise due to the lower porosity and permeability in hard rock settings, which result in limited and complex pathways for fluid movement. The Mont Terri underground rock laboratory in Switzerland provides the opportunity to evaluate different geophysical measurements in boreholes to monitor a CO<span><math><msub><mrow></mrow><mrow><mn>2</mn></mrow></msub></math></span> injection near a known fault zone. The motivation of the experiment presented in this study, along with its continuation in the Carbon Sequestration series D and E (CS-D and CS-E) is to assess the integrity of a fault zone within a caprock-like formation (such as Opalinus Clay) during long-term leakage experiment. Time-lapse electrical resistivity tomography (ERT) measurements were conducted during the steady-state injection period. In this work we present an ERT study using synthetic data to predict the effects of both the conductive fault zone and the injected fluid mixture at a single time step, while also analyzing the temporal evolution of the synthetic study. The synthetic results show a similar apparent resistivity distribution to that observed in the field data. The analysis then progressed to real field data, where various electrode configurations were tested, requiring meticulous assessment of data quality during processing. This study highlights the importance of using appropriate error estimation techniques, such as a reciprocal error model, to characterize the spatial and temporal behavior of measurement errors across different configurations. Three-dimensional time-lapse inversion results play a crucial role in deciphering the fluid interaction between the injected CO<span><math><msub><mrow></mrow><mrow><mn>2</mn></mrow></msub></math></span>, the properties of the host rock, and the presence of the main fault zone within the experiment. Our findings indicate that the fracture network within the host rock is intricate, exhibiting changes in resistivity during injection around the main fault zone. These insights not only complement other findings within the CS-D and CS-E experiments, but also showcase the utility of ERT measurements in CO<span><math><msub><mrow></mrow><mrow><mn>2</mn></mrow></msub></math></span> monitoring within other hard rock settings.</div></div>","PeriodicalId":54882,"journal":{"name":"Journal of Applied Geophysics","volume":"242 ","pages":"Article 105852"},"PeriodicalIF":2.1,"publicationDate":"2025-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144766726","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Heterogeneous oil reservoirs characterization using artificial intelligence techniques: Application to the Hassi Messaoud oil field in the Algerian-Saharan platform","authors":"Asma Kadri ,&nbsp;Mohammed Said Benzagouta","doi":"10.1016/j.jappgeo.2025.105878","DOIUrl":"10.1016/j.jappgeo.2025.105878","url":null,"abstract":"<div><div>This study explores the effectiveness of artificial intelligence (AI) techniques in characterizing heterogenous reservoirs, with a specific focus on the Hassi Messaoud oil field in southern Algeria, particularly its newly developed northern zone known for complex reservoir heterogeneity affecting oil extraction efficiency.</div><div>Three machine learning algorithms Support Vector Machine (SVM), Random Forest (RF), and Artificial Neural Networks (ANN) were applied to predict porosity from well log data. SVM outperformed RF and ANN, delivering the highest correlation coefficients (R²) and the lowest root mean squared errors (RMSE), thereby confirming its robustness in high-dimensional spaces and limited datasets. Beyond porosity prediction, this study addresses the critical task of permeability estimation, essential for optimizing reservoir development strategies. A Multi-Linear Regression (MLR) model was developed, achieving high predictive accuracy, particularly when lithological parameters such as clay content were incorporated. A key contribution of this work is the integration of SVM and MLR into a hybrid SVM–MLR model, which improved permeability prediction by leveraging the strengths of both methods: the nonlinear feature-handling capability of SVM and the interpretability of MLR. The SVM-generated correlation matrix facilitated the identification of dominant input features, enhancing the reliability of the permeability model.</div><div>The findings demonstrate that the integrated SVM–MLR approach provides a powerful and adaptable tool for reservoir characterization in heterogeneous environments. This AI-driven framework offers valuable support for data-informed decision-making, contributing to more efficient hydrocarbon recovery and improved reservoir management in geologically complex settings.</div></div>","PeriodicalId":54882,"journal":{"name":"Journal of Applied Geophysics","volume":"242 ","pages":"Article 105878"},"PeriodicalIF":2.1,"publicationDate":"2025-07-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144750234","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Analysis of sensitive clays from the Gjerdrum landslide in Norway: Dielectric spectroscopy with modeled TEM IP-responses, and x-ray diffraction analysis","authors":"R. Skartlien ,&nbsp;E. Finden ,&nbsp;K.D. Knudsen ,&nbsp;J.I. Loshelder ,&nbsp;T.M.H. Le ,&nbsp;S. Tavakoli ,&nbsp;S.E. Hamran ,&nbsp;T. Vestgården ,&nbsp;T.L. Palmer","doi":"10.1016/j.jappgeo.2025.105871","DOIUrl":"10.1016/j.jappgeo.2025.105871","url":null,"abstract":"<div><div>Dielectric spectra of pure and silty clays of varying sensitivity from the Romerike clay deposit were obtained to model induced polarization (IP) signatures in TEM (transient electromagnetics) data obtained with stationary current loops on the ground. The Maxwell–Wagner (MW) relaxation frequencies that characterizes the dielectric spectra were in the range 6–11 kHz, and the results indicate higher relaxation frequencies for more sensitive clay. X-ray diffraction analysis showed nearly identical mineral composition of all the clay samples, with chlorite and illite as the main components. The results indicate that both the relaxation frequency and clay sensitivity increase with silt contents. Although pore water conductivity (due to salt and other ions) did not correlate well to sensitivity and relaxation frequency in the current samples, it is still likely that the ionic composition influences these parameters. Relaxation frequency variation in clay layers a few meters below the surface can be detected with TEM provided that the dielectric strength of the clay is sufficiently large (above about 1500), and that the resistivity of the overlying soil layer is sufficiently large (above about 100 <span><math><mrow><mi>Ω</mi><mi>m</mi></mrow></math></span>). Hence, TEM could be used to determine subsurface clay sensitivity under the right conditions, once the link between the relaxation frequency and clay sensitivity is fully understood.</div></div>","PeriodicalId":54882,"journal":{"name":"Journal of Applied Geophysics","volume":"242 ","pages":"Article 105871"},"PeriodicalIF":2.1,"publicationDate":"2025-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144780020","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Spatiotemporal evolution patterns of mining-induced overburden damage based on microseismic event response analysis","authors":"Xiao Yang ,&nbsp;Longqing Shi ,&nbsp;Mei Qiu ,&nbsp;Jin Han ,&nbsp;Xingyue Qu ,&nbsp;Song Fu","doi":"10.1016/j.jappgeo.2025.105876","DOIUrl":"10.1016/j.jappgeo.2025.105876","url":null,"abstract":"<div><div>Jurassic coal mining in the Ordos Basin faces threats from overlying water-bearing sandstone layers of the Yijun and Luohe Formations, posing challenges for predicting the dynamic evolution prediction of spatial relationships between water-conducting fracture zones and aquifers. This study integrated rock beam theoretical models with microseismic monitoring data to analyze the spatiotemporal evolution of overburden fracture characteristics during longwall face retreat. Through mechanical model analysis of the overburden, we elucidated the stress-driving mechanism of individual rock beams in the mining direction (transverse), thereby establishing a longitudinal “four-zone” structural model of overburden fracture (Collapsed water-conducting zone, Rock-beam water-conducting zone, Damaged rock-beam weak-permeability zone, and Initial water-resisting zone), identifying the upper boundary of water-conducting fracture zone as the damaged rock-beam weak-permeability zone. Taking Longwall Face 1502 of Shaozhai Coal Mine as the research object, spatiotemporal analysis of microseismic data during production revealed that microseismic frequency time series exhibited a fluctuating decreasing trend and dynamically responded to rock beam fracture characteristics. Based on this, we applied singular spectrum analysis coupled with a distance-weighted energy algorithm to process microseismic data, identifying significant overburden fracture changes at 74, 197, 320, and 432 days of face advancement. By integrating coal measures depositional environments, microseismic event density, and stratigraphic columns, we dynamically determined the overburden fracture boundary and maximum development of the water-conducting fracture zone extended to the 37.46 m mudstone layer of the Anding Formation. Validation through water injection tests confirmed that the Yijun and Luohe Formations remained within the Initial water-resisting zone, undisturbed by mining activities. The ternary analysis method of “stress field driving → microseismic response → sedimentary constraint” proposed in this study reveals the dynamic correlation between mining-induced overburden fracture and microseismic responses, providing a theoretical foundation for effective water prevention and control measures.</div></div>","PeriodicalId":54882,"journal":{"name":"Journal of Applied Geophysics","volume":"242 ","pages":"Article 105876"},"PeriodicalIF":2.1,"publicationDate":"2025-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144722190","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Integrated petrophysical characterization of Late Cretaceous fluvial-deltaic channel systems in Penobscot Field, Nova Scotia: Implications for reservoir facies prediction","authors":"Satya Narayan ,&nbsp;Pradeep Kumar ,&nbsp;Soumyashree Debasis Sahoo ,&nbsp;Birendra Pratap ,&nbsp;Ahmed M. Eldosouky","doi":"10.1016/j.jappgeo.2025.105877","DOIUrl":"10.1016/j.jappgeo.2025.105877","url":null,"abstract":"<div><div>Globally, deltaic channels offer promising exploration prospects as potential hydrocarbon reservoirs. Sea-level changes in deltaic systems cause facies heterogeneity in and around channels, making precise reservoir characterization essential to reduce exploration risk. This study focuses on the petrophysical characterization of the deltaic channel system within the Logan Canyon Formation in the Penobscot Field. The workflow integrates structural mapping, reservoir feasibility assessment, spectral decomposition, model-based inversion (MBI), and multi-layer perceptron (MLP) modeling. Litho-facies, acoustic impedance, and petrophysical property relationships were quantified through well-log cross plots. Subsequently, impedance and petrophysical volumes were accurately predicted using MBI and MLP techniques. The results show an 84.7 % correlation with a synthetic error of ∼0.12 and impedance error of 414 m/s*g/cm³ for P-imp, and an 89.6 % correlation with an error of ∼0.026 for the Vsh, 95.5 % correlation with an error of ∼0.013 for Nphi, and 86.1 % correlation with an error of ∼0.017 for Phie estimations. Sand reservoir facies show moderate to higher P-imp (8000–10,500 m/s*g/cm³), lower Vsh (&lt;0.40 <em>v</em>/v), and lower Nphi (0.25–0.35 <em>v</em>/v), with higher Phie (0.12 to 0.30 <em>v</em>/v) signatures, whereas shale facies exhibit moderate to lower P-imp (6500–9000 m/s*g/cm³), higher Vsh (&gt;0.40 <em>v</em>/v), higher Nphi (&gt;0.30 v/v), and lower Phie (0.0 to 0.05 v/v) signatures. This facies differentiation highlights the spatial heterogeneity within the deltaic system and delineates the reservoir zones within complex channel architectures. Integration of P-imp, Nphi, Vsh, and Phie provides a robust geological model that enhances the understanding of facies dispersal pattern and reservoir potential, and identifies sweet spots for future hydrocarbon exploration within the Late Cretaceous channel system of the Penobscot Field, Nova Scotia.</div></div>","PeriodicalId":54882,"journal":{"name":"Journal of Applied Geophysics","volume":"242 ","pages":"Article 105877"},"PeriodicalIF":2.1,"publicationDate":"2025-07-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144738430","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Imaging near surface formation using TEM and ERT methods: A case study from the western coast of India","authors":"Bhima Raju Kattula ,&nbsp;Subash Chandra","doi":"10.1016/j.jappgeo.2025.105875","DOIUrl":"10.1016/j.jappgeo.2025.105875","url":null,"abstract":"<div><div>The paper presents comparative performance of airborne electromagnetic (EM), ground-based EM and electrical resistivity tomography (ERT) methods for characterization of near-surface hydrogeological units of the Western coast of India. The electromagnetic data was acquired in time domain mode employing dual (low and high) moments technology, while the ERT data was obtained using multi-gradient arrays with 10 m electrode spacing. The early time gates of low moment (LM) data provide near-surface geological knowledge, whereas late time gate measurements of high moment (HM) data provide deeper information. Early time gate measurements of LM data corrected by primary field and system response help in utilizing a very early time gate starting from ∼4 μs that enhances near surface resolution. Formation resistivity derived from LM airborne transient electromagnetic (aTEM) data has been compared with ground-based transient electromagnetic (gTEM) and ERT data for resolving very near-surface features, especially the upper few meters, including soil. The study revealed that the performance of aTEM with system response corrections and on-time modelling is effective in comparison to gTEM, and consistent with those results from ERT of 10 m unit electrode spacing in resolving near-surface features. Thus, aTEM survey can contribute to hydrogeological applications dealing with shallow processes such as infiltration, and aquifer recharge studies.</div></div>","PeriodicalId":54882,"journal":{"name":"Journal of Applied Geophysics","volume":"242 ","pages":"Article 105875"},"PeriodicalIF":2.2,"publicationDate":"2025-07-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144704078","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Depth structural mapping of syn and pre-rift rock units of southeast Ghara at southern end of Gulf of Suez","authors":"Osama Elghrabawy ,&nbsp;Shaimaa Abd El Raheem ,&nbsp;Alaa Aref ,&nbsp;Khaled Khalifa","doi":"10.1016/j.jappgeo.2025.105874","DOIUrl":"10.1016/j.jappgeo.2025.105874","url":null,"abstract":"<div><div>The interpretation of seismic data in the Gulf of Suez area for identifying and imaging pre-Miocene horizons is problematic due to the influence of thick evaporites near the surface. Consequently, this paper sheds light on the imaging of shallow and deep structure complexity of southeast Ghara based on an integrated analysis of seismic and aeromagnetic data. The analysis strategy included in-depth structural mapping of syn and pre-rift formations. The results indicate that the Miocene and pre-Miocene structure complexities are related to southwest-dipping blocks controlled by major listric faults associated with dip-slip faults. These structures are considered the main cause of forming the tilted faulted blocks in the study area, with a trend NNW to NW and many cross faults in ENE and NE in the study area with NNW to NW trends and many cross faults in the ENE and NE directions. One of these tilted faulted blocks forms the main horst structure trend in the Ghara area. The depth structure maps show that the depths of formations increase gradually westward and decrease eastward. The sub-basins in the western part exhibit an average thickness of 5029 m and are bounded by a salt ridge in the southeast and the uplifted offshore basement in the northwest. The lowermost portions of the Miocene formations and the pre-rift strata thickness are identified by mapping the basement and the base of the Rudeis Formation, which exhibit an average thickness of 549 m. The structure complexity analysis shows probable locations to entrap and accumulate hydrocarbons. The interpretation suggests that the up-thrown side of the interpreted normal faults show promising locations for hydrocarbon accumulations, especially along the central to southern, and northern portions of the Ghara study area.</div></div>","PeriodicalId":54882,"journal":{"name":"Journal of Applied Geophysics","volume":"241 ","pages":"Article 105874"},"PeriodicalIF":2.2,"publicationDate":"2025-07-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144662163","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Integration of TDEM and HVSR surveys for water resources investigation in arid regions: The case of Horombe plateau, Madagascar","authors":"Arianna Paschetto ,&nbsp;Andrea Vergnano ,&nbsp;Chiara Caselle ,&nbsp;Cesare Comina ,&nbsp;Luigi Benente ,&nbsp;Edoardo Bono ,&nbsp;Taddeo Maria Fenoglio ,&nbsp;Francesca Fontanella ,&nbsp;Davide Vianello ,&nbsp;Sabrina Maria Rita Bonetto","doi":"10.1016/j.jappgeo.2025.105872","DOIUrl":"10.1016/j.jappgeo.2025.105872","url":null,"abstract":"<div><div>The increasing threat of droughts, due to climate variability and rising water demands, is further exacerbated by the lack of hydrogeological knowledge of remote areas such as Southern Madagascar. This study aims to showcase the utility of cost-effective and portable geophysical methods alongside with detailed geological and hydrogeological investigations to enhance water resource exploration in arid regions.</div><div>For this purpose, integration of Horizontal-to-Vertical Spectral Ratio (HVSR) and Time Domain Electromagnetic Method (TDEM) has been adopted to investigate a 5 km² area, with 41 survey points, in a zone of the Horombe plateau, in South Madagascar, remote and mostly unknown from the geological point of view.</div><div>Considering the dominant frequencies of HVSR, the TDEM resistivity model and the few available literature and field geological data (a water well sampling campaign), a geological interpretation of geophysical data was performed to assess the distribution and extent of water resources. The proposed geological/hydrogeological interpretation consists of a laterite cover, hosting irregular and ephemeral aquifers, a saprolite layer, hosting the main aquifer, and a gneiss bedrock.</div><div>According to the results of this study, three new water wells were drilled where geophysical data detected the thickest aquifer. The wells validated the proposed geological model and confirmed the presence of main aquifers in the bottom part of saprolite layer and thinner aquifers in the lateritic cover.</div><div>The results of the study offer a framework for the application of easy-to-operate geophysical solutions for water resources management in remote arid areas with scarcity of available data. The paper also discusses the limitations and uncertainties in geophysical data interpretation, particularly regarding the achievable investigation depth.</div></div>","PeriodicalId":54882,"journal":{"name":"Journal of Applied Geophysics","volume":"241 ","pages":"Article 105872"},"PeriodicalIF":2.2,"publicationDate":"2025-07-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144694280","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"TLAD-YOLO: Lightweight network for intelligent detection of railway tunnel lining anomalies using ground penetrating radar","authors":"Pengfei Zheng ,&nbsp;Anxue Zhang ,&nbsp;Zhensheng Shi ,&nbsp;Sen Wang ,&nbsp;Yi'an Ma ,&nbsp;Zhaodan Liu","doi":"10.1016/j.jappgeo.2025.105869","DOIUrl":"10.1016/j.jappgeo.2025.105869","url":null,"abstract":"<div><div>Ground Penetrating Radar (GPR) B-scan images and the you only look once (YOLO) series are widely used for tunnel lining intelligent inspections to ensure quality. However, in practical applications, lightweight YOLO detection networks fail to meet the requirements of accuracy and robustness. In view of this, a tunnel lining anomalies detection YOLO (TLAD-YOLO) is proposed for the intelligent detection of railway tunnel lining anomalies based on GPR B-scan images. TLAD-YOLO introduces lightweight spatial and channel synergistic multi-shape attention (SCSMSA) to enhance the detection accuracy of complex scenes and multi-size abnormal objects, while ghost convolution is used to reduce parameters and computation. The experiments are conducted on a dataset consisting of 47 railway tunnels. Furthermore, we propose a multi-scale data augmentation to further expand the dataset, which improves the detection accuracy. The experimental results demonstrate that TLAD-YOLO is an accurate and lightweight detection network, outperforming SOTA detection networks in non-destructive testing of railway tunnels. On the tunnel engineering verification platform and newly built railway tunnels, TLAD-YOLO demonstrates remarkable robustness.</div></div>","PeriodicalId":54882,"journal":{"name":"Journal of Applied Geophysics","volume":"241 ","pages":"Article 105869"},"PeriodicalIF":2.2,"publicationDate":"2025-07-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144662154","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}