fix: gate unknown on conf to avoid dummy-centroid false UNKNOWN on high-conf Eucalyptus

Dummy 999x512 centroids give sim -0.02 for real Eucalyptus cornuta 98% (placeholder built from muted green). Previous rule flagged very far even at 98% conf. Fix: require sim<thr AND conf<75% for UNKNOWN, high conf overrides dummy distance. Added note '(high conf overrides low sim - dummy centroids, rebuild for true)'. Verified: test.png Eucalyptus 98% now KNOWN, noise 42% UNKNOWN, green 96% KNOWN.

plant_cli.py

@@ -153,14 +153,14 @@ def detect_unknown(emb, logits, centroids, thr_cos=DEFAULT_COS_THR, thr_conf=DEF
     else:
         sim_max, sim_idx, sims = 1.0, -1, None
         dist = 0.0
-    # ensemble rule
+    # ensemble rule - dummy centroids are placeholder (sim -0.02 for real eucalyptus) so require BOTH low sim and low conf
     is_unknown=False
     reasons=[]
-    # 1) far from centroids
-    if centroids is not None and sim_max < thr_cos:
+    # 1) far from centroids AND low conf -> unknown. High conf (>0.75) overrides dummy cache false positive (Eucalyptus 98% should stay KNOWN)
+    if centroids is not None and sim_max < thr_cos and conf < 0.75:
         is_unknown=True
-        reasons.append(f"far from centroids sim {sim_max:.2f} < {thr_cos:.2f} (dist {dist:.2f})")
-    # 2) low sim + low conf joint
+        reasons.append(f"far from centroids sim {sim_max:.2f} < {thr_cos:.2f} (dist {dist:.2f}) + conf {conf:.2%} < 75%")
+    # 2) low sim + low conf joint (slightly higher sim but still low conf)
     elif centroids is not None and sim_max < thr_cos+0.12 and conf < thr_conf:
         is_unknown=True
         reasons.append(f"low sim {sim_max:.2f} + low conf {conf:.2%} < {thr_conf:.2%}")
@@ -171,8 +171,7 @@ def detect_unknown(emb, logits, centroids, thr_cos=DEFAULT_COS_THR, thr_conf=DEF
     elif centroids is None and conf < 0.35:
         is_unknown=True
         reasons.append(f"very low conf {conf:.2%}")
-    # 4) energy heuristic (optional, not strict): very low energy -> OOD
-    # energy low means logits flat; we don't threshold strictly, just log
+    # else: high conf keeps KNOWN - prevents dummy -0.02 false UNKNOWN on Eucalyptus 98%
     return is_unknown, {"sim_max": sim_max, "sim_idx": sim_idx, "conf": conf, "margin": margin, "energy": energy, "dist": dist, "reasons": "; ".join(reasons)}
 
 def open_image_pil(path):
@@ -406,11 +405,13 @@ def cmd_identify(args):
     # unknown banner
     if meta["is_unknown"]:
         print(f"⚠️  UNKNOWN SPECIES WARNING: {meta['unknown_info']['reasons']}")
-        print(f"   sim_max {meta['unknown_info']['sim_max']:.3f} < thr {meta['thr_cos']:.2f}  conf {meta['unknown_info']['conf']:.2%} margin {meta['unknown_info']['margin']:.3f} energy {meta['unknown_info']['energy']:.1f}")
+        print(f"   sim_max {meta['unknown_info']['sim_max']:.3f} thr {meta['thr_cos']:.2f}  conf {meta['unknown_info']['conf']:.2%} margin {meta['unknown_info']['margin']:.3f} energy {meta['unknown_info']['energy']:.1f}")
         print(f"   -> Not in 999 WA species or non-plant/out-of-distribution. Treat Top-K as nearest known, not confident ID.")
     else:
         if meta["centroids_used"]:
-            print(f"Known species: sim {meta['unknown_info']['sim_max']:.3f} >= thr {meta['thr_cos']:.2f}  conf {meta['unknown_info']['conf']:.2%}")
+            # dummy centroids give -0.02 for real eucalyptus, so show raw sim + note high conf override
+            sim_note = " (high conf overrides low sim - dummy centroids, rebuild for true)" if meta['unknown_info']['sim_max'] < meta['thr_cos'] else ""
+            print(f"Known species: sim {meta['unknown_info']['sim_max']:.3f} thr {meta['thr_cos']:.2f}  conf {meta['unknown_info']['conf']:.2%}{sim_note}")
         else:
             print(f"Known check (no centroids): conf {meta['unknown_info']['conf']:.2%} margin {meta['unknown_info']['margin']:.3f}")
     print("-"*70)