Add detection operator profiles

CHANGELOG.md

@@ -7,6 +7,13 @@
 
 # Changelog
 
+## Sprint 155 Detection operator profiles (2026-07-09)
+
+- Added explicit Detection Lab operator profiles for the inactive `geointel-building-yolov8s-aoi1024bg512r3e50-pt` local model asset.
+- Added a balanced review profile at confidence threshold `0.15` and a conservative review profile at `0.35`, with persisted gate metrics shown in the UI.
+- Kept both profiles clearly marked as candidate-only and not default-approved because the promotion recommendation remains `none` and background false-positive pressure still blocks automatic activation.
+- No model download behavior, API contract, migration, provider fetching, fake detection output or active runtime default changed.
+
 ## Sprint 154 Background-aware AOI1024 YOLOv8s candidate gate (2026-07-09)
 
 - Exported and audited background-aware AOI1024 training dataset `/app/storage/operator-data/yolo-building-aoi1024-bgaware512r3`; the audit passed with 162 tiles, 117 positive tiles, 45 negative tiles, 21,530 labels and no warnings.

backend/tests/test_sprint122_model_asset_activation_guardrails.py

@@ -20,10 +20,9 @@ def test_detection_lab_explains_explicit_model_asset_and_threshold_selection() -
 
     assert "Explicit model asset" in lab
     assert "No model file is selected automatically" in lab
-    assert "Current benchmark candidate" in lab
-    assert "geointel-building-yolov8s-hardneg160r4e50-pt" in lab
-    assert "Recommended starting threshold" in lab
-    assert "0.25" in lab
+    assert "Operator profiles" in lab
+    assert "DETECTION_OPERATOR_PROFILES" in lab
+    assert "Candidate only - not default-approved" in lab
     assert "will_download_models" in lab
 
 

backend/tests/test_sprint155_detection_operator_profiles.py

@@ -0,0 +1,46 @@
+from pathlib import Path
+
+
+ROOT = Path(__file__).resolve().parents[2]
+
+
+def test_detection_operator_profiles_define_explicit_non_default_yolo_candidates() -> None:
+    profiles = ROOT / "frontend" / "src" / "components" / "detection" / "detectionProfiles.ts"
+    source = profiles.read_text(encoding="utf-8")
+
+    assert "DETECTION_OPERATOR_PROFILES" in source
+    assert "geointel-building-yolov8s-aoi1024bg512r3e50-pt" in source
+    assert "balanced-review" in source
+    assert "conservative-review" in source
+    assert "confidenceThreshold: 0.15" in source
+    assert "confidenceThreshold: 0.35" in source
+    assert "defaultApproved: false" in source
+    assert "promotionRecommendation: 'none'" in source
+    assert "false-positive pressure" in source
+
+
+def test_detection_lab_surfaces_profiles_as_deliberate_operator_actions() -> None:
+    lab = (ROOT / "frontend" / "src" / "components" / "detection" / "DetectionLab.tsx").read_text(
+        encoding="utf-8"
+    )
+
+    assert "DETECTION_OPERATOR_PROFILES" in lab
+    assert "Operator profiles" in lab
+    assert "profile.displayName" in lab
+    assert "profile.confidenceThreshold" in lab
+    assert "Candidate only - not default-approved" in lab
+    assert "Apply profile" in lab
+    assert "onApplyOperatorProfile(profile)" in lab
+    assert "Recommended starting threshold: 0.25" not in lab
+
+
+def test_detection_workflow_applies_profiles_without_auto_selecting_assets() -> None:
+    hook = (ROOT / "frontend" / "src" / "hooks" / "useDetectionWorkflow.ts").read_text(encoding="utf-8")
+    app = (ROOT / "frontend" / "src" / "App.tsx").read_text(encoding="utf-8")
+
+    assert "applyDetectionOperatorProfile" in hook
+    assert "setSelectedDetectionModelId('yolo-configured')" in hook
+    assert "setSelectedModelAssetId(profile.modelAssetId)" in hook
+    assert "setDetectionConfidenceThreshold(profile.confidenceThreshold)" in hook
+    assert "setSelectedModelAssetId(assetResponse.items[0]" not in hook
+    assert "onApplyOperatorProfile={applyDetectionOperatorProfile}" in app

docs/AI_PIPELINES.md

@@ -260,6 +260,14 @@ as false-positive pressure. It does not run QA/QC or invent reference metrics
 for empty/sparse background AOIs. The first expanded local model improved dense
 AOI F1, but Kasterlee-bos false positives block default promotion.
 
+The current inactive AOI1024 background-aware local model asset,
+`geointel-building-yolov8s-aoi1024bg512r3e50-pt`, is exposed in Detection Lab
+only through deliberate operator profiles. `balanced-review` applies threshold
+`0.15` for the strongest positive-AOI F1 observed so far; `conservative-review`
+applies threshold `0.35` for higher precision review. Both profiles remain
+candidate-only, not default-approved, because the promotion recommendation is
+still `none` and background false-positive pressure has not passed the gate.
+
 To compare the same model/tile/threshold grid across all prepared operator
 samples, use:
 

docs/CODEX_EXECUTION_LOG.md

@@ -6168,3 +6168,32 @@ Open:
 
 - Build a V1 operator detection profile layer in the UI/docs: expose `balanced` (`threshold=0.15`) and `conservative review` (`threshold=0.35`) as explicit choices for local model assets, with clear warning that the model is not a default-approved detector.
 - Clean the background corpus classification: separate pure-empty AOIs from sparse-building contextual AOIs, then retrain or recalibrate against that cleaner gate.
+
+# Sprint 155 - Detection operator profiles
+
+## What changed
+
+- Added `frontend/src/components/detection/detectionProfiles.ts` with explicit operator profiles for the inactive `geointel-building-yolov8s-aoi1024bg512r3e50-pt` local model asset.
+- Exposed two deliberate Detection Lab actions:
+  - `balanced-review`: confidence threshold `0.15`, positive-AOI F1 `0.5074022485589402`, precision `0.636639`, recall `0.424258`, max background detections `103`.
+  - `conservative-review`: confidence threshold `0.35`, positive-AOI F1 `0.32086574003576274`, precision `0.840006`, recall `0.202135`, max background detections `55`.
+- Applying a profile selects `yolo-configured`, the local model asset id and the profile threshold. It does not auto-select assets on model catalog load and does not promote the candidate as a default detector.
+- Detection Lab now marks both profiles as `Candidate only - not default-approved` because the promotion recommendation remains `none`.
+- Updated frontend, AI pipeline and TODO documentation.
+
+## What was tested
+
+- Added regression coverage in `backend/tests/test_sprint155_detection_operator_profiles.py`.
+- Ran `python -m pytest tests/test_sprint155_detection_operator_profiles.py tests/test_sprint122_model_asset_activation_guardrails.py -q`.
+- Ran `python -m pytest` in `backend`: 435 passed.
+- Ran `python -m compileall backend/app`.
+- Ran `cd frontend && npm run typecheck`.
+- Ran `cd frontend && npm run build`.
+- Ran `bash scripts/run_readiness_check.sh`.
+- Ran `cd backend && python -m alembic heads` and `cd backend && python -m alembic upgrade head --sql`.
+- Ran `bash -n scripts/live_migration_smoke.sh`.
+
+## Known limitations
+
+- The profiles are review/demo aids only. The background corpus still needs to be split into pure-empty negatives and sparse-building contextual AOIs before retraining or recalibrating for a default detector decision.
+- No backend API contract, migration, provider fetching, fake detection output, model download behavior or active runtime default changed.

docs/TODO.md

@@ -120,7 +120,7 @@ This file now starts with the current implementation status. Older preparation/b
 - [x] Export and audit AOI1024 clean-label variants; select `yolo-building-aoi1024-visible050-minpx8` as the first audit-passing 512px training candidate.
 - [x] Train and gate `geointel-building-yolov8s-aoi1024clean512e50-pt` through seven positive AOIs and nine hard-negative/background AOIs.
 - [x] Train and gate background-aware `geointel-building-yolov8s-aoi1024bg512r3e50-pt`; it is the strongest positive-AOI candidate so far but remains inactive because full background-candidate false-positive pressure still blocks default promotion.
-- [ ] Add explicit operator detection profiles for local model assets: balanced review around threshold `0.15` and conservative high-precision review around threshold `0.35`, both clearly marked as non-default-approved until promotion gates pass.
+- [x] Add explicit operator detection profiles for local model assets: balanced review around threshold `0.15` and conservative high-precision review around threshold `0.35`, both clearly marked as non-default-approved until promotion gates pass.
 - [ ] Split the background corpus into pure-empty negatives and sparse-building contextual AOIs, then retrain or recalibrate against the cleaner gate.
 - [ ] Promote a V1 default building detector only after it passes seven positive AOIs, clean hard-negative/background gates and persisted QA/QC evidence without fake detections or model downloads.
 

frontend/README.md

@@ -122,6 +122,7 @@ AI Lab run controls explicitly explain when no raster dataset is available, inst
 - Detection Lab now exposes the `yolo-configured` capability reported by the backend.
 - When `yolo-configured` is selected, users can provide an existing raster tile manifest path.
 - Detection Lab lists local model assets from `GET /api/v1/detection/model-assets` so operators can choose an existing mounted model file instead of editing only one hidden `YOLO_MODEL_PATH` slot.
+- Detection Lab exposes explicit operator profiles for the current inactive local AOI1024 building detector: balanced review at threshold `0.15` and conservative review at threshold `0.35`. Applying a profile deliberately selects the local model asset and threshold; it does not approve or promote a default model.
 - Detection Lab includes a read-only YOLO runtime preflight panel with backend status, dependency visibility, local model configuration, `torch`/`ultralytics` versions, CUDA state and `YOLO_CONFIG_DIR`.
 - The UI still does not download models or create fake detections; backend status and error codes remain the source of truth.
 

frontend/src/App.tsx

@@ -260,6 +260,7 @@ function App(): JSX.Element {
     runDetection,
     runDetectionQa,
     runDetectionCalibration,
+    applyDetectionOperatorProfile,
     resetDetectionForProject,
     setSelectedDetectionDatasetId,
     setSelectedDetectionModelId,
@@ -994,6 +995,7 @@ function App(): JSX.Element {
                 onSetCalibrationThresholdText={setCalibrationThresholdText}
                 onRunCalibration={runDetectionCalibration}
                 onOpenCalibrationEvidence={openQualityEvidenceOnMap}
+                onApplyOperatorProfile={applyDetectionOperatorProfile}
               />
 
               <SegmentationLab

frontend/src/components/detection/DetectionLab.tsx

@@ -10,6 +10,7 @@ import type {
   YoloPreflightResponse,
 } from '../../types'
 import type { DetectionCalibrationRunRow } from '../../hooks/useDetectionWorkflow'
+import { DETECTION_OPERATOR_PROFILES, type DetectionOperatorProfile } from './detectionProfiles'
 
 interface CalibrationRow {
   analysisRunId: string
@@ -80,6 +81,7 @@ interface DetectionLabProps {
   onSetCalibrationThresholdText: (value: string) => void
   onRunCalibration: () => void
   onOpenCalibrationEvidence: (qualityCheckId: string) => void
+  onApplyOperatorProfile: (profile: DetectionOperatorProfile) => void
 }
 
 export function DetectionLab({
@@ -135,6 +137,7 @@ export function DetectionLab({
   onSetCalibrationThresholdText,
   onRunCalibration,
   onOpenCalibrationEvidence,
+  onApplyOperatorProfile,
 }: DetectionLabProps): JSX.Element {
   const selectedDetectionModel = detectionModels.find((model) => model.model_id === selectedDetectionModelId) ?? null
   const selectedModelAsset = modelAssets.find((asset) => asset.model_asset_id === selectedModelAssetId) ?? null
@@ -145,9 +148,6 @@ export function DetectionLab({
   const detectionModelUiRunnable = detectionModelReady && selectedDetectionModelId !== 'manual-fixture-detector'
   const detectionHasExplicitModelAsset =
     selectedDetectionModelId !== 'yolo-configured' || modelAssets.length === 0 || selectedModelAssetId.length > 0
-  const benchmarkCandidateAsset = modelAssets.find(
-    (asset) => asset.model_asset_id === 'geointel-building-yolov8s-hardneg160r4e50-pt',
-  )
   const calibrationRows = buildCalibrationRows(detectionRuns, qualityChecks)
   const bestF1Candidate = bestCalibrationRow(calibrationRows, 'f1')
   const bestPrecisionCandidate = bestCalibrationRow(calibrationRows, 'precision')
@@ -251,12 +251,62 @@ export function DetectionLab({
               {selectedModelAsset ? 'asset selected' : 'no explicit asset'}
             </span>
           </div>
-          {benchmarkCandidateAsset ? (
+          <div className="model-asset-guidance">
+            <strong>Operator profiles</strong>
+            <p>
+              Candidate profiles apply a local model asset and confidence threshold only after an explicit click.
+              Candidate only - not default-approved while the promotion recommendation remains none.
+            </p>
+          </div>
+          <div className="operator-profile-grid" aria-label="Configured YOLO operator profiles">
+            {DETECTION_OPERATOR_PROFILES.map((profile) => {
+              const profileAsset = modelAssets.find((asset) => asset.model_asset_id === profile.modelAssetId)
+              const profileSelected =
+                selectedModelAssetId === profile.modelAssetId &&
+                Math.abs(detectionConfidenceThreshold - profile.confidenceThreshold) < 0.0001
+              return (
+                <div
+                  className={profileSelected ? 'operator-profile-card operator-profile-card-selected' : 'operator-profile-card'}
+                  key={profile.id}
+                >
+                  <div className="operator-profile-card-header">
+                    <strong>{profile.displayName}</strong>
+                    <span className={profile.defaultApproved ? 'status-badge status-badge-ready' : 'status-badge'}>
+                      {profile.defaultApproved ? 'default-approved' : 'Candidate only - not default-approved'}
+                    </span>
+                  </div>
+                  <p>{profile.description}</p>
+                  <div className="operator-profile-metrics">
+                    <span>threshold {profile.confidenceThreshold.toFixed(2)}</span>
+                    <span>precision {profile.precision.toFixed(3)}</span>
+                    <span>recall {profile.recall.toFixed(3)}</span>
+                    <span>F1 {profile.f1.toFixed(3)}</span>
+                    <span>max background FP {profile.maxBackgroundDetections}</span>
+                  </div>
+                  <div className="entity-meta">
+                    <span>asset: {profile.modelAssetId}</span>
+                    <span>promotionRecommendation: {profile.promotionRecommendation}</span>
+                    <span>available: {profileAsset ? 'yes' : 'not mounted'}</span>
+                  </div>
+                  <p className="field-guidance">{profile.limitationMessage}</p>
+                  <button
+                    className="secondary-action"
+                    type="button"
+                    onClick={() => onApplyOperatorProfile(profile)}
+                    disabled={!profileAsset}
+                  >
+                    Apply profile
+                  </button>
+                </div>
+              )
+            })}
+          </div>
+          {selectedModelAsset ? (
             <div className="model-asset-guidance">
-              <strong>Current benchmark candidate</strong>
+              <strong>Selected model asset status</strong>
               <p>
-                {benchmarkCandidateAsset.display_name} is available for deliberate evaluation. Recommended starting threshold: 0.25.
-                Keep it operator-selected until hard-negative false positives are reduced.
+                {selectedModelAsset.display_name} is operator-selected. Keep local candidates inactive until persisted
+                promotion evidence explicitly recommends default activation.
               </p>
             </div>
           ) : null}
@@ -469,7 +519,7 @@ export function DetectionLab({
               />
               {selectedDetectionModelId === 'yolo-configured' ? (
                 <span className="field-guidance">
-                  Recommended starting threshold: 0.25 for the current local YOLOv8s benchmark candidate.
+                  Use an operator profile for the current local YOLOv8s candidate, or enter a threshold manually for calibration.
                 </span>
               ) : null}
             </label>

frontend/src/components/detection/detectionProfiles.ts

@@ -0,0 +1,47 @@
+export interface DetectionOperatorProfile {
+  id: string
+  displayName: string
+  modelAssetId: string
+  confidenceThreshold: number
+  defaultApproved: boolean
+  promotionRecommendation: 'none' | 'promote_candidate'
+  precision: number
+  recall: number
+  f1: number
+  maxBackgroundDetections: number
+  description: string
+  limitationMessage: string
+}
+
+export const DETECTION_OPERATOR_PROFILES: DetectionOperatorProfile[] = [
+  {
+    id: 'balanced-review',
+    displayName: 'Balanced review',
+    modelAssetId: 'geointel-building-yolov8s-aoi1024bg512r3e50-pt',
+    confidenceThreshold: 0.15,
+    defaultApproved: false,
+    promotionRecommendation: 'none',
+    precision: 0.636639,
+    recall: 0.424258,
+    f1: 0.5074022485589402,
+    maxBackgroundDetections: 103,
+    description: 'Best positive-AOI F1 profile for deliberate operator review of the inactive AOI1024 model asset.',
+    limitationMessage:
+      'Candidate only because false-positive pressure still blocks default promotion on the background/hard-negative gate.',
+  },
+  {
+    id: 'conservative-review',
+    displayName: 'Conservative review',
+    modelAssetId: 'geointel-building-yolov8s-aoi1024bg512r3e50-pt',
+    confidenceThreshold: 0.35,
+    defaultApproved: false,
+    promotionRecommendation: 'none',
+    precision: 0.840006,
+    recall: 0.202135,
+    f1: 0.32086574003576274,
+    maxBackgroundDetections: 55,
+    description: 'Higher-precision profile for demos or review sessions where fewer false positives matter more than recall.',
+    limitationMessage:
+      'Candidate only because false-positive pressure remains visible; use it deliberately and inspect persisted QA evidence.',
+  },
+]

frontend/src/hooks/useDetectionWorkflow.ts

@@ -21,6 +21,11 @@ interface DetectionWorkflowOptions {
   loadQualityChecks: (projectId?: string | null) => Promise<QualityCheckRead[] | void>
 }
 
+interface DetectionOperatorProfileSelection {
+  modelAssetId: string
+  confidenceThreshold: number
+}
+
 export interface DetectionCalibrationRunRow {
   threshold: number
   status: 'queued' | 'running' | 'success' | 'failed'
@@ -333,6 +338,12 @@ export function useDetectionWorkflow({
     }
   }
 
+  const applyDetectionOperatorProfile = (profile: DetectionOperatorProfileSelection) => {
+    setSelectedDetectionModelId('yolo-configured')
+    setSelectedModelAssetId(profile.modelAssetId)
+    setDetectionConfidenceThreshold(profile.confidenceThreshold)
+  }
+
   const resetDetectionForProject = () => {
     setSelectedDetectionDatasetId('')
     setDetectionRuns([])
@@ -383,6 +394,7 @@ export function useDetectionWorkflow({
     runDetection,
     runDetectionQa,
     runDetectionCalibration,
+    applyDetectionOperatorProfile,
     resetDetectionForProject,
     setSelectedDetectionDatasetId,
     setSelectedDetectionModelId,

frontend/src/styles/app.css

@@ -3473,6 +3473,67 @@ button.entity-card {
   line-height: 1.35;
 }
 
+.operator-profile-grid {
+  display: grid;
+  grid-template-columns: repeat(auto-fit, minmax(16rem, 1fr));
+  gap: 0.62rem;
+}
+
+.operator-profile-card {
+  display: grid;
+  gap: 0.48rem;
+  min-width: 0;
+  border: 1px solid #d8e3de;
+  border-radius: 8px;
+  padding: 0.72rem;
+  background: #ffffff;
+}
+
+.operator-profile-card-selected {
+  border-color: var(--accent);
+  background: #f7fffc;
+  box-shadow: 0 0 0 3px rgba(15, 118, 110, 0.1);
+}
+
+.operator-profile-card-header {
+  display: flex;
+  min-width: 0;
+  align-items: flex-start;
+  justify-content: space-between;
+  gap: 0.6rem;
+}
+
+.operator-profile-card-header strong {
+  min-width: 0;
+  color: var(--text);
+  font-size: 0.94rem;
+  line-height: 1.25;
+}
+
+.operator-profile-card p {
+  margin: 0;
+  color: var(--muted);
+  font-size: 0.8rem;
+  line-height: 1.35;
+}
+
+.operator-profile-metrics {
+  display: grid;
+  grid-template-columns: repeat(auto-fit, minmax(6.4rem, 1fr));
+  gap: 0.36rem;
+}
+
+.operator-profile-metrics span {
+  border: 1px solid #e0e9e4;
+  border-radius: 6px;
+  padding: 0.34rem 0.42rem;
+  background: #f9fbfa;
+  color: var(--text);
+  font-size: 0.76rem;
+  font-weight: 700;
+  overflow-wrap: anywhere;
+}
+
 .field-guidance {
   display: block;
   margin-top: 0.24rem;