{"id":"W4396897807","doi":"10.7717/peerj-cs.2003","title":"Applying a deep learning pipeline to classify land cover from low-quality historical RGB imagery","year":2024,"lang":"en","type":"article","venue":"PeerJ Computer Science","topic":"Remote-Sensing Image Classification","field":"Engineering","cited_by":5,"is_retracted":false,"has_abstract":true,"ca_institutions":"","funders":"Environment and Climate Change Canada","keywords":"Land cover; Satellite imagery; Cover (algebra); Remote sensing; Land use; Convolutional neural network; Computer science; Deep learning; Artificial intelligence; Physical geography; Geography; Ecology; Engineering","routes":{"ca_aff":false,"ca_fund":true,"ca_venue":false,"about_ca":true,"invisible_to_affiliation_only":true},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0007320772,0.0001857521,0.0002169341,0.0002439504,0.0001595808,0.0006072811,0.0003865779,0.00006046869,0.00002239031],"category_scores_gemma":[0.0001715833,0.0001837641,0.0000642422,0.0009400753,0.00009253358,0.0004669971,0.0001657354,0.0003759998,0.000663876],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0008485711,"about_ca_system_score_gemma":0.00006409834,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00008603513,"about_ca_topic_score_gemma":0.000005780099,"domain_scores_codex":[0.9980317,0.00004598103,0.0003278372,0.0006148539,0.0005886458,0.0003909512],"domain_scores_gemma":[0.9989977,0.0002524448,0.00003057997,0.0003931535,0.0001226763,0.000203463],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.000009109926,0.00003137047,0.0009108019,0.0001332002,0.00001640105,0.0000813786,0.001411123,0.2174081,0.1866379,0.00008467373,0.01814779,0.5751282],"study_design_scores_gemma":[0.00007221517,0.00001152558,0.003903702,0.00006436399,0.000006614787,0.000009015083,0.000003056389,0.9291649,0.001772626,0.00005469597,0.06470704,0.0002302879],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.04746174,0.0003315673,0.9472027,0.0006637833,0.002544145,0.0001759767,0.000002856414,0.0009244064,0.0006927756],"genre_scores_gemma":[0.9072531,0.00001154701,0.09108693,0.0002490801,0.0007586046,0.0000133296,0.0000104191,0.00003938323,0.0005775495],"genre_candidate":"methods","genre_consensus":null,"teacher_disagreement_score":0.8597914,"threshold_uncertainty_score":0.8533004,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.02246771798820442,"score_gpt":0.2646689674354333,"score_spread":0.2422012494472289,"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."}}