{"id":"W4408391585","doi":"10.1016/j.ndteint.2025.103366","title":"Singular-value Gain Compensation: Robust and efficient GPR preprocessing method enhancing zero-shot underground object segmentation by Segment-Anything Model","year":2025,"lang":"en","type":"article","venue":"NDT & E International","topic":"Geophysical Methods and Applications","field":"Engineering","cited_by":4,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Fusion Oriented REsearch for disruptive Science and Technology; Japan Science and Technology Agency; Council for Science, Technology and Innovation; Swine Innovation Porc","keywords":"Preprocessor; Segmentation; Compensation (psychology); Artificial intelligence; Object (grammar); Computer vision; Ground-penetrating radar; Zero (linguistics); Shot (pellet); Computer science; Pattern recognition (psychology); Engineering; Mathematics; Materials science; Radar; Psychology","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0003969537,0.0001780992,0.0001694653,0.0001114411,0.0001924065,0.0002151212,0.0001552997,0.00007244282,0.00002329926],"category_scores_gemma":[0.00006401668,0.0001958814,0.00005178546,0.0001956658,0.00003163158,0.000174125,0.00006891062,0.000193813,0.000005353077],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0002435805,"about_ca_system_score_gemma":0.00003914302,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00003578655,"about_ca_topic_score_gemma":0.000005637731,"domain_scores_codex":[0.9987902,0.00005711302,0.0003415033,0.0003435414,0.000278522,0.000189104],"domain_scores_gemma":[0.999378,0.0002437074,0.00007301924,0.0001526981,0.00009836668,0.00005417977],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000005152086,0.00004640669,0.00004253273,0.00005012243,0.00006322333,3.644828e-7,0.0003328297,0.6780314,0.3008702,0.008744868,0.0002147658,0.01159808],"study_design_scores_gemma":[0.0003239107,0.000006683128,0.0002766492,0.0001120805,0.00003469131,0.000003124944,0.0001655667,0.9493155,0.04079463,0.008582517,0.0002164367,0.0001681821],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1073294,0.0001359086,0.8874839,0.0003985369,0.0002522612,0.0002002153,0.00001782031,0.0001297463,0.004052266],"genre_scores_gemma":[0.7136621,0.00001064441,0.2855096,0.000196135,0.00005665288,0.00006069397,0.00007727202,0.0000194563,0.0004073738],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.6063328,"threshold_uncertainty_score":0.7987811,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.01979691295746829,"score_gpt":0.3123894691289316,"score_spread":0.2925925561714633,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}