{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:FLFOQOLVC7PEHBFAZCCQ7X7IJB","short_pith_number":"pith:FLFOQOLV","canonical_record":{"source":{"id":"2604.27617","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-04-30T09:08:41Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c26977d1713e6fdc85dbae7996a8fad5b3a3b59630f2d91c98ffc2c85eaca0b3","abstract_canon_sha256":"259fb75bfd695badf4071735ad3ff35e902462b5163125ff863b251557a161bc"},"schema_version":"1.0"},"canonical_sha256":"2acae8397517de4384a0c8850fdfe8484b53d98ba14673d7a2309175af3ae0ba","source":{"kind":"arxiv","id":"2604.27617","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.27617","created_at":"2026-06-01T01:02:40Z"},{"alias_kind":"arxiv_version","alias_value":"2604.27617v2","created_at":"2026-06-01T01:02:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.27617","created_at":"2026-06-01T01:02:40Z"},{"alias_kind":"pith_short_12","alias_value":"FLFOQOLVC7PE","created_at":"2026-06-01T01:02:40Z"},{"alias_kind":"pith_short_16","alias_value":"FLFOQOLVC7PEHBFA","created_at":"2026-06-01T01:02:40Z"},{"alias_kind":"pith_short_8","alias_value":"FLFOQOLV","created_at":"2026-06-01T01:02:40Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:FLFOQOLVC7PEHBFAZCCQ7X7IJB","target":"record","payload":{"canonical_record":{"source":{"id":"2604.27617","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-04-30T09:08:41Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"c26977d1713e6fdc85dbae7996a8fad5b3a3b59630f2d91c98ffc2c85eaca0b3","abstract_canon_sha256":"259fb75bfd695badf4071735ad3ff35e902462b5163125ff863b251557a161bc"},"schema_version":"1.0"},"canonical_sha256":"2acae8397517de4384a0c8850fdfe8484b53d98ba14673d7a2309175af3ae0ba","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-01T01:02:40.938593Z","signature_b64":"JgKDVnZrG5qODvSa5i5hESl93e51GefzKFDoS+uA/pof7H4Gr8ZkWzLOM79X6485WCRaxELJoBWoUXiaoFbDCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2acae8397517de4384a0c8850fdfe8484b53d98ba14673d7a2309175af3ae0ba","last_reissued_at":"2026-06-01T01:02:40.937274Z","signature_status":"signed_v1","first_computed_at":"2026-06-01T01:02:40.937274Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2604.27617","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-01T01:02:40Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"w5uuUhakMi1j0Wb3QlVnxHVr6HQvR5WjDoEDyVlpk0EQyQRJ+g3U6RIIhQXlWqSlnF9iy67jDdDo6K+HerRiCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-04T15:12:57.449529Z"},"content_sha256":"88eaba1d70a01e9a064b9868b707f62ab857ebc1116b889e0c372ab80344f570","schema_version":"1.0","event_id":"sha256:88eaba1d70a01e9a064b9868b707f62ab857ebc1116b889e0c372ab80344f570"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:FLFOQOLVC7PEHBFAZCCQ7X7IJB","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Robust Lightweight Crack Classification for Real-Time UAV Bridge Inspection","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"Lightweight CNN with attention and focal loss detects bridge cracks at 825 FPS for UAV inspections.","cross_cats":["cs.AI"],"primary_cat":"cs.CV","authors_text":"Haisheng Li, Jiandong Wang, Kaichen Ma, Luming Yang, Weijie Li, Wei Li","submitted_at":"2026-04-30T09:08:41Z","abstract_excerpt":"With the widespread application of Unmanned Aerial Vehicles (UAVs) in bridge structural health monitoring, deep learning-based automatic crack detection has become a major research focus. However, practical UAV inspections still face four key challenges: weak crack features, degraded imaging conditions, severe class imbalance, and limited computational resources for practical UAV inspection workflows. To address these issues, this paper proposes a unified lightweight convolutional neural network framework composed of four synergistic components: a lightweight backbone network, a Convolutional "},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"Experiments on the SDNET2018 bridge deck dataset show that the proposed method achieves an inference speed of 825 FPS with only 11.21M parameters and 1.82G FLOPs. Compared with the baseline model, the complete framework improves the F1-score by 2.51% and recall by 3.95%.","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That the directed robust augmentation strategy based on inspection-scene priors, when combined with CBAM and Focal Loss, will produce robust performance gains that generalize from the SDNET2018 dataset to varied real-world UAV flight conditions, lighting, and bridge types not represented in the test set.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A lightweight CNN framework with CBAM, targeted augmentation, and focal loss runs at 825 FPS on the SDNET2018 dataset while improving F1-score by 2.51% and recall by 3.95% over baseline for bridge crack detection.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Lightweight CNN with attention and focal loss detects bridge cracks at 825 FPS for UAV inspections.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"49683dd6064760bb194b00637ea9e2b6091b668302b797732e72daee4d95da5a"},"source":{"id":"2604.27617","kind":"arxiv","version":2},"verdict":{"id":"582d2aa2-e1b2-441c-b5b4-472a95246c47","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-07T05:25:43.786227Z","strongest_claim":"Experiments on the SDNET2018 bridge deck dataset show that the proposed method achieves an inference speed of 825 FPS with only 11.21M parameters and 1.82G FLOPs. Compared with the baseline model, the complete framework improves the F1-score by 2.51% and recall by 3.95%.","one_line_summary":"A lightweight CNN framework with CBAM, targeted augmentation, and focal loss runs at 825 FPS on the SDNET2018 dataset while improving F1-score by 2.51% and recall by 3.95% over baseline for bridge crack detection.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That the directed robust augmentation strategy based on inspection-scene priors, when combined with CBAM and Focal Loss, will produce robust performance gains that generalize from the SDNET2018 dataset to varied real-world UAV flight conditions, lighting, and bridge types not represented in the test set.","pith_extraction_headline":"Lightweight CNN with attention and focal loss detects bridge cracks at 825 FPS for UAV inspections."},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2604.27617/integrity.json","findings":[],"available":true,"detectors_run":[{"name":"ai_meta_artifact","ran_at":"2026-05-20T21:43:28.372071Z","status":"completed","version":"1.0.0","findings_count":0},{"name":"doi_compliance","ran_at":"2026-05-19T19:03:57.676469Z","status":"completed","version":"1.0.0","findings_count":0}],"snapshot_sha256":"1e919a2b3e0850efa0a185f82de710dd3b621328c2f7d8d220e261081ffe83cc"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"verdict_id":"582d2aa2-e1b2-441c-b5b4-472a95246c47"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-06-01T01:02:40Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"GpAlnCCn3gUh+KqBswRKUSORAvFUlzXYeT/JeAIIVgR0zslyZ/q35zQ8uhfMg6oF74cAn0pwgRtqFL2wFAGQCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-04T15:12:57.450058Z"},"content_sha256":"f590a6958cf90cc5a25cbbd718c999232672b29c715bded741b0b3dff1867230","schema_version":"1.0","event_id":"sha256:f590a6958cf90cc5a25cbbd718c999232672b29c715bded741b0b3dff1867230"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/FLFOQOLVC7PEHBFAZCCQ7X7IJB/bundle.json","state_url":"https://pith.science/pith/FLFOQOLVC7PEHBFAZCCQ7X7IJB/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/FLFOQOLVC7PEHBFAZCCQ7X7IJB/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-06-04T15:12:57Z","links":{"resolver":"https://pith.science/pith/FLFOQOLVC7PEHBFAZCCQ7X7IJB","bundle":"https://pith.science/pith/FLFOQOLVC7PEHBFAZCCQ7X7IJB/bundle.json","state":"https://pith.science/pith/FLFOQOLVC7PEHBFAZCCQ7X7IJB/state.json","well_known_bundle":"https://pith.science/.well-known/pith/FLFOQOLVC7PEHBFAZCCQ7X7IJB/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:FLFOQOLVC7PEHBFAZCCQ7X7IJB","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"259fb75bfd695badf4071735ad3ff35e902462b5163125ff863b251557a161bc","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-04-30T09:08:41Z","title_canon_sha256":"c26977d1713e6fdc85dbae7996a8fad5b3a3b59630f2d91c98ffc2c85eaca0b3"},"schema_version":"1.0","source":{"id":"2604.27617","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2604.27617","created_at":"2026-06-01T01:02:40Z"},{"alias_kind":"arxiv_version","alias_value":"2604.27617v2","created_at":"2026-06-01T01:02:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2604.27617","created_at":"2026-06-01T01:02:40Z"},{"alias_kind":"pith_short_12","alias_value":"FLFOQOLVC7PE","created_at":"2026-06-01T01:02:40Z"},{"alias_kind":"pith_short_16","alias_value":"FLFOQOLVC7PEHBFA","created_at":"2026-06-01T01:02:40Z"},{"alias_kind":"pith_short_8","alias_value":"FLFOQOLV","created_at":"2026-06-01T01:02:40Z"}],"graph_snapshots":[{"event_id":"sha256:f590a6958cf90cc5a25cbbd718c999232672b29c715bded741b0b3dff1867230","target":"graph","created_at":"2026-06-01T01:02:40Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":4,"items":[{"attestation":"unclaimed","claim_id":"C1","kind":"strongest_claim","source":"verdict.strongest_claim","status":"machine_extracted","text":"Experiments on the SDNET2018 bridge deck dataset show that the proposed method achieves an inference speed of 825 FPS with only 11.21M parameters and 1.82G FLOPs. Compared with the baseline model, the complete framework improves the F1-score by 2.51% and recall by 3.95%."},{"attestation":"unclaimed","claim_id":"C2","kind":"weakest_assumption","source":"verdict.weakest_assumption","status":"machine_extracted","text":"That the directed robust augmentation strategy based on inspection-scene priors, when combined with CBAM and Focal Loss, will produce robust performance gains that generalize from the SDNET2018 dataset to varied real-world UAV flight conditions, lighting, and bridge types not represented in the test set."},{"attestation":"unclaimed","claim_id":"C3","kind":"one_line_summary","source":"verdict.one_line_summary","status":"machine_extracted","text":"A lightweight CNN framework with CBAM, targeted augmentation, and focal loss runs at 825 FPS on the SDNET2018 dataset while improving F1-score by 2.51% and recall by 3.95% over baseline for bridge crack detection."},{"attestation":"unclaimed","claim_id":"C4","kind":"headline","source":"verdict.pith_extraction.headline","status":"machine_extracted","text":"Lightweight CNN with attention and focal loss detects bridge cracks at 825 FPS for UAV inspections."}],"snapshot_sha256":"49683dd6064760bb194b00637ea9e2b6091b668302b797732e72daee4d95da5a"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[{"findings_count":0,"name":"ai_meta_artifact","ran_at":"2026-05-20T21:43:28.372071Z","status":"completed","version":"1.0.0"},{"findings_count":0,"name":"doi_compliance","ran_at":"2026-05-19T19:03:57.676469Z","status":"completed","version":"1.0.0"}],"endpoint":"/pith/2604.27617/integrity.json","findings":[],"snapshot_sha256":"1e919a2b3e0850efa0a185f82de710dd3b621328c2f7d8d220e261081ffe83cc","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"With the widespread application of Unmanned Aerial Vehicles (UAVs) in bridge structural health monitoring, deep learning-based automatic crack detection has become a major research focus. However, practical UAV inspections still face four key challenges: weak crack features, degraded imaging conditions, severe class imbalance, and limited computational resources for practical UAV inspection workflows. To address these issues, this paper proposes a unified lightweight convolutional neural network framework composed of four synergistic components: a lightweight backbone network, a Convolutional ","authors_text":"Haisheng Li, Jiandong Wang, Kaichen Ma, Luming Yang, Weijie Li, Wei Li","cross_cats":["cs.AI"],"headline":"Lightweight CNN with attention and focal loss detects bridge cracks at 825 FPS for UAV inspections.","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-04-30T09:08:41Z","title":"Robust Lightweight Crack Classification for Real-Time UAV Bridge Inspection"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2604.27617","kind":"arxiv","version":2},"verdict":{"created_at":"2026-05-07T05:25:43.786227Z","id":"582d2aa2-e1b2-441c-b5b4-472a95246c47","model_set":{"reader":"grok-4.3"},"one_line_summary":"A lightweight CNN framework with CBAM, targeted augmentation, and focal loss runs at 825 FPS on the SDNET2018 dataset while improving F1-score by 2.51% and recall by 3.95% over baseline for bridge crack detection.","pipeline_version":"pith-pipeline@v0.9.0","pith_extraction_headline":"Lightweight CNN with attention and focal loss detects bridge cracks at 825 FPS for UAV inspections.","strongest_claim":"Experiments on the SDNET2018 bridge deck dataset show that the proposed method achieves an inference speed of 825 FPS with only 11.21M parameters and 1.82G FLOPs. Compared with the baseline model, the complete framework improves the F1-score by 2.51% and recall by 3.95%.","weakest_assumption":"That the directed robust augmentation strategy based on inspection-scene priors, when combined with CBAM and Focal Loss, will produce robust performance gains that generalize from the SDNET2018 dataset to varied real-world UAV flight conditions, lighting, and bridge types not represented in the test set."}},"verdict_id":"582d2aa2-e1b2-441c-b5b4-472a95246c47"}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:88eaba1d70a01e9a064b9868b707f62ab857ebc1116b889e0c372ab80344f570","target":"record","created_at":"2026-06-01T01:02:40Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"259fb75bfd695badf4071735ad3ff35e902462b5163125ff863b251557a161bc","cross_cats_sorted":["cs.AI"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2026-04-30T09:08:41Z","title_canon_sha256":"c26977d1713e6fdc85dbae7996a8fad5b3a3b59630f2d91c98ffc2c85eaca0b3"},"schema_version":"1.0","source":{"id":"2604.27617","kind":"arxiv","version":2}},"canonical_sha256":"2acae8397517de4384a0c8850fdfe8484b53d98ba14673d7a2309175af3ae0ba","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"2acae8397517de4384a0c8850fdfe8484b53d98ba14673d7a2309175af3ae0ba","first_computed_at":"2026-06-01T01:02:40.937274Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-06-01T01:02:40.937274Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"JgKDVnZrG5qODvSa5i5hESl93e51GefzKFDoS+uA/pof7H4Gr8ZkWzLOM79X6485WCRaxELJoBWoUXiaoFbDCw==","signature_status":"signed_v1","signed_at":"2026-06-01T01:02:40.938593Z","signed_message":"canonical_sha256_bytes"},"source_id":"2604.27617","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:88eaba1d70a01e9a064b9868b707f62ab857ebc1116b889e0c372ab80344f570","sha256:f590a6958cf90cc5a25cbbd718c999232672b29c715bded741b0b3dff1867230"],"state_sha256":"63521c176c1ccc1934b89eba3e0eae2fe5256b1017261b74f241bc0c0b48d2a2"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"oOmO1Pr6BbvCBXebkHKta9nm2JvLEzwqeScocaBzLcyk2y2uRicIvZBgj2O9nTSnEHO18OPTofAh2qU89QFvCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-04T15:12:57.452487Z","bundle_sha256":"cf762ac76073f9e51e5fa6e15c23c4b1d5269a4979e3449633f7c14c6337749a"}}