Harden operator tile background metadata

scripts/export_operator_yolo_tile_dataset.py

@@ -21,6 +21,9 @@ from typing import Any, Iterable
 
 DEFAULT_MANIFEST_PATH = Path("/app/storage/operator-data/operator_samples_manifest.json")
 DEFAULT_OUTPUT_DIR = Path("/app/storage/operator-data/yolo-building-tile-dataset")
+REFERENCE_AOI_CATEGORY = "reference_aoi"
+PURE_EMPTY_BACKGROUND_CATEGORY = "pure_empty_negative"
+SPARSE_BACKGROUND_CATEGORY = "sparse_building_context"
 rasterio: Any = None
 Window: Any = None
 Transformer: Any = None
@@ -165,6 +168,18 @@ def background_negative_repeat_count(
     return max(1, background_negative_repeat)
 
 
+def background_category_for_sample(sample: dict[str, Any]) -> str:
+    explicit_category = str(sample.get("background_category") or "").strip()
+    if explicit_category:
+        return explicit_category
+    if str(sample.get("sample_role") or "reference") != "background_candidate":
+        return REFERENCE_AOI_CATEGORY
+    reference_feature_count = int(sample.get("reference_feature_count") or 0)
+    if reference_feature_count <= 0:
+        return PURE_EMPTY_BACKGROUND_CATEGORY
+    return SPARSE_BACKGROUND_CATEGORY
+
+
 def resolve_manifest_path(raw: str, manifest_path: Path) -> Path:
     path = Path(raw)
     if path.exists():
@@ -315,7 +330,7 @@ def export_sample_tiles(
 ) -> list[dict[str, Any]]:
     sample_slug = str(sample["sample_slug"])
     sample_role = str(sample.get("sample_role") or "reference")
-    background_category = str(sample.get("background_category") or "reference_aoi")
+    background_category = background_category_for_sample(sample)
     split = "val" if sample_slug.lower() in val_slugs else "train"
     raster_path = resolve_manifest_path(str(sample["raster_path"]), manifest_path)
     reference_path = resolve_manifest_path(str(sample["reference_path"]), manifest_path)