IBBI-bio/ibbi_yolov11_oc

Model Card: yolov11x_bb_multi_class_detect_model

Model Details

• Model Name: yolov11x_bb_multi_class_detect_model
• Model Type: Multi-Class Object Detection and Classifier
• Description: This model is designed to detect and classify specific species and genera of bark beetles from images. Unlike single-class models, it has been fine-tuned on a labeled dataset of bark beetle species.

---

Evaluation Datasets

To understand the model's capabilities, its performance was tested on two different types of datasets:

• In-Distribution (ID): This dataset contains images of species the model was trained on. Performance on this dataset shows how well the model identifies familiar species.
• Out-of-Distribution (OOD): This dataset contains images of species that are intentionally different from the training data. Performance here tests the model's ability to handle novel species.

---

Performance

Object Detection & Classification

The model's performance is measured by its mean Average Precision (mAP). This score reflects the model's accuracy in both locating the beetle (bounding box) and assigning the correct species or genus label.

Species-Level Performance

This evaluates the model's ability to identify individual species.

Dataset	Species mAP (0.50 : 0.95)	Notes
In-Distribution (ID)	🟩 0.9002	Excellent overall accuracy, with strong performance on most trained species.
Out-of-Distribution (OOD)	🟥 0.0000	As expected, the model cannot classify species it has not been trained on.

Click to see Per-Species Performance (ID Dataset)

The following list is sorted by Average Precision (AP) from lowest to highest to highlight the most challenging species for the model to identify.

Species	AP Score
Scolytus_multistriatus	0.0222
Dendroctonus_rufipennis	0.1667
Euwallacea_validus	0.2222
Hylesinus_aculeatus	0.3626
Ips_grandicollis	0.4074
Dryocoetes_autographus	0.5500
Trypodendron_domesticum	0.6875
Xyleborus_celsus	0.7069
Orthotomicus_caelatus	0.7284
Ambrosiodmus_minor	0.8017
Ambrosiophilus_atratus	0.9049
Hylurgus_ligniperda	0.9361
Pityogenes_chalcographus	0.9367
Xylosandrus_germanus	0.9600
Ips_typographus	0.9659
Coccotrypes_dactyliperda	0.9671
Scolytus_schevyrewi	0.9672
Ips_sexdentatus	0.9678
Monarthrum_mali	0.9690
Dendroctonus_terebrans	0.9695
Xylosandrus_crassiusculus	0.9707
Xyleborus_ferrugineus	0.9718
Xyleborinus_saxesenii	0.9732
Xylosandrus_compactus	0.9777
Monarthrum_fasciatum	0.9791
Euplatypus_compositus	0.9828
Pagiocerus_frontalis	0.9829
Xylosandrus_amputatus	0.9826
Hypothenemus_hampei	0.9822
Cnestus_mutilatus	0.9853
Cryptocarenus_heveae	0.9854
Hylesinus_varius	0.9859
Ips_calligraphus	0.9862
Ips_avulsus	0.9877
Orthotomicus_erosus	0.9879
Cyclorhipidion_pelliculosum	0.9886
Dendroctonus_valens	0.9898
Hylesinus_toranio	0.9890
Scolytodes_glaber	0.9896
Hylurgops_palliatus	0.9901
Pityophthorus_juglandis	0.9902
Hylastes_porculus	0.9902
Xyleborus_glabratus	0.9905
Xyleborus_affinis	0.9916
Euwallacea_fornicatus	0.9918
Coccotrypes_carpophagus	0.9921
Myoplatypus_flavicornis	0.9947
Ctonoxylon_hagedorn	0.9958
Euwallacea_perbrevis	0.9968
Platypus_cylindrus	0.9979
Phloeosinus_dentatus	0.9980
Tomicus_destruens	0.9995
Anisandrus_sayi	0.9998
Xylosandrus_morigerus	0.9998
Ips_duplicatus	0.9998
Pycnarthrum_hispidium	0.9999
Coptoborus_ricini	0.9999
Anisandrus_dispar	0.9581
Ips_acuminatus	1.0000
Platypus_koryoensis	1.0000
Hylesinus_crenatus	1.0000
Hylastes_salebrosus	1.0000

---

Genus-Level Performance

This evaluates the model's ability to identify the genus, a broader taxonomic rank than species.

Dataset	Genus mAP (0.50 : 0.95)	Notes
In-Distribution (ID)	🟩 0.9446	Excellent performance on genera the model was trained to recognize.
Out-of-Distribution (OOD)	🟩 0.7741	Shows strong generalization, successfully classifying many unseen genera with good accuracy.

Click to see Per-Genus Performance (ID and OOD Datasets)

The following lists are sorted by Average Precision (AP) from lowest to highest to highlight the most challenging genera for the model to identify.

In-Distribution (ID) Genus Performance

Genus	AP Score
Dryocoetes	0.7000
Hypothenemus	0.8866
Trypodendron	0.8875
Pityogenes	0.8938
Scolytus	0.9000
Cryptocarenus	0.9070
Xylosandrus	0.9113
Taphrorychus	0.9126
Coptoborus	0.9180
Pityophthorus	0.9213
Xyleborinus	0.9294
Scolytodes	0.9302
Coccotrypes	0.9326
Euwallacea	0.9415
Pycnarthrum	0.9467
Pagiocerus	0.9498
Orthotomicus	0.9529
Xyleborus	0.9555
Monarthrum	0.9565
Ambrosiophilus	0.9585
Anisandrus	0.9629
Hylurgus	0.9697
Cnestus	0.9735
Ips	0.9754
Euplatypus	0.9755
Cyclorhipidion	0.9773
Phloeosinus	0.9779
Ctonoxylon	0.9803
Dendroctonus	0.9832
Hylesinus	0.9846
Hylurgops	0.9865
Platypus	0.9909
Ambrosiodmus	0.9909
Hylastes	0.9922
Myoplatypus	0.9958
Tomicus	0.9980

Out-of-Distribution (OOD) Genus Performance

Genus	AP Score
Dactylotrypes	0.2583
Cryptocarenus	0.3832
Cryphalus	0.3869
Pityogenes	0.4214
Crypturgus	0.4846
Carphoborus	0.5185
Hylocurus	0.5189
Polygraphus	0.5222
Premnobius	0.6095
Leptoxyleborus	0.6294
Dendroctonus	0.6638
Pycnarthrum	0.6750
Dinoplatypus	0.6833
Ernoporus	0.6859
Cyclorhipidion	0.7087
Hypothenemus	0.7114
Crossotarsus	0.7176
Wallacellus	0.7400
Pityoborus	0.7556
Diuncus	0.7587
Trypodendron	0.7591
Webbia	0.7733
Xylocleptes	0.7872
Dendroterus	0.7867
Monarthrum	0.7929
Gnathotrichus	0.7926
Cnestus	0.7937
Truncaudum	0.8121
Scolytus	0.8172
Hylastes	0.8218
Ambrosiodmus	0.8233
Xyloterinus	0.8393
Chaetoptelius	0.8382
Xyleborus	0.8461
Eccoptopterus	0.8444
Debus	0.8500
Heteroborips	0.8533
Xyleborinus	0.8548
Euwallacea	0.8608
Ips	0.8682
Hadrodemius	0.8688
Microperus	0.8756
Tomicus	0.8829
Eidophelus	0.8838
Coptoborus	0.8929
Pseudopityophthorus	0.8970
Metacorthylus	0.9000
Platypus	0.9143
Hylurgus	0.9196
Anisandrus	0.9160
Procryphalus	0.9231
Pityophthorus	0.9401
Dryocoetes	0.9444
Beaverium	0.9545
Tricosa	0.9615
Stegomerus	0.9842
Pseudowebbia	0.9917
Cnesinus	1.0000

---

Feature Extraction (Embedding Performance)

The quality of the model's learned feature representations (embeddings) is evaluated by how well they group similar species together.

Internal Cluster Validation

These metrics measure the quality of the clusters formed by the embeddings without referring to ground-truth labels.

Metric	ID Score	OOD Score	Interpretation
Silhouette Score	0.7606	0.2305	Measures how similar an object is to its own cluster compared to others. Higher is better (closer to 1). The ID embeddings form exceptional, well-defined clusters.
Davies-Bouldin Index	0.3303	0.4092	Measures the average similarity between each cluster and its most similar one. Lower is better (closer to 0). The ID embeddings show very little overlap.
Calinski-Harabasz Index	14511.8	948.297	Measures the ratio of between-cluster dispersion to within-cluster dispersion. Higher is better. The ID embeddings form exceptionally dense and well-separated clusters.

External Cluster Validation

These metrics evaluate the clustering performance by comparing the results to the true species labels.

Metric	ID Score	OOD Score	Interpretation
Adjusted Rand Index (ARI)	0.3519	0.0071	Measures the similarity between true and predicted labels, correcting for chance. Higher is better (closer to 1).
Normalized Mutual Info (NMI)	0.6865	0.3254	Measures the agreement between the clustering and the true labels. Higher is better (closer to 1).
Cluster Purity	0.6646	0.1656	Measures the extent to which clusters contain a single class. Higher is better (closer to 1).

Conclusion: The high external validation scores for the ID dataset show that the model's feature representations are effective at separating the different species it was trained on. This is a significant improvement over single-class or zero-shot models.

Phylogenetic Correlation (Mantel Test)

This test determines if the model's learned features correlate with the evolutionary relationships (phylogeny) between different bark beetle species.

Dataset	Mantel R-statistic	p-value	Interpretation
In-Distribution (ID)	-0.1517	0.1860	There is no statistically significant correlation between the model's features and the species' evolutionary history.
Out-of-Distribution (OOD)	-0.0068	0.9310	There is no statistically significant correlation for the OOD data either.