CropCop: An Auditable 120-Class Plant-Health Model from Benchmark Reconstruction to a Quantised Runtime Artifact

Rana Muhammad Ahmed; Sabahat Abbas

cs.CVAug 26, 2026

CropCop: An Auditable 120-Class Plant-Health Model from Benchmark Reconstruction to a Quantised Runtime Artifact

Authors: Rana Muhammad Ahmed, Sabahat Abbas

Organizations: Department of Computer Science, Bahria University Islamabad, Islamabad, Pakistan

Abstract

A plant-health score can appear precise while resting on duplicated image families, a long-tailed label space, or a runtime file that was never evaluated. We present CropCop, a closed-set recognition system spanning 120 operational plant-health classes and an evidence chain from corpus reconstruction to direct execution of the final quantised artifact. Starting from 117,546 audited images, we rejected the inherited partition after confirming 3,233 duplicate relationships across split boundaries and froze a 109,107-image benchmark with zero crossings among the audited trusted leakage groups and a 151.7 largest-to-smallest class ratio. A fully fine-tuned DINOv3 ConvNeXt-Tiny reference achieved 98.51% accuracy and 96.87% macro-F1 on the locked internal test. A compact MobileNetV4 Conv-Medium derivative achieved 98.46% accuracy and 96.27% macro-F1 without being presented as evidence for a new distillation method. Validation-only post-training quantisation selected dynamic activations with per-channel weights, and the final 22.60 MiB ExecuTorch/XNNPACK PTE achieved 98.46% accuracy and 96.23% macro-F1 when executed directly. Only six of 16,363 top-1 decisions changed between the converted INT8 graph and the PTE, while paired analysis showed a modest class-balanced loss; an exploratory post hoc fruit-label slice localized a larger recall decline than aggregate accuracy revealed. CropCop establishes strong leakage-controlled internal recognition and software-runtime fidelity; it does not establish performance on unseen farms, camera pipelines, or physical Android hardware.

Figures & tables

Figure 1: CropCop audit-to-runtime evidence chain.

Literature line	Primary object	What it establishes	Gap relative to CropCop’s question
Controlled and in-the-wild plant benchmarks [ 1 , 2 , 3 , 4 , 5 , 6 , 7 ]	Dataset and recognition task	Scale, taxonomy, acquisition difficulty, or localisation	The benchmark is normally the endpoint; compact runtime identity is not the central object.
Benchmark audit and reporting [ 8 , 9 , 10 , 11 , 12 , 13 ]	Contamination, labels, shortcuts, documentation	Why data identity, provenance, and limitations matter	These works are domain-general and do not trace one agricultural model through quantised execution.
Foundation transfer and long-tail learning [ 14 , 15 , 16 , 17 , 18 ]	Representations and class imbalance	Transfer baselines and class-balanced objectives	They do not by themselves establish benchmark integrity or runtime fidelity.
Distillation and efficient vision [ 19 , 20 , 21 , 22 , 23 , 24 ]	Compact-model training and quantisation	How accuracy can be transferred or compressed	Dataset forensics and exact serialized-artifact evaluation are usually separate concerns.
Calibration and runtime tooling [ 25 , 26 , 27 ]	Confidence and executable backends	How predictions can be calibrated, exported, lowered, and quantised	They supply methods and infrastructure rather than an end-to-end application evidence chain.
CropCop	One audited plant-health model lineage	Benchmark reconstruction, locked evaluation, compact transfer, PTQ selection, direct PTE execution, and artifact identity	The contribution is the connected evidence chain, not a new backbone and not a claim beyond the evaluated distribution.

Table 1: Positioning of CropCop relative to adjacent literature.

Dimension	Category	Count
Crop scope	Named crop or plant groups	35
Crop scope	Crop-unspecified labels	2
Label type	Specific named disease †	72
Label type	Broad condition category	9
Label type	Generic healthy state	39
Label type	Total operational classes	120

Table 2: Operational class-map composition.

Figure 2: Historical contamination and the frozen replacement.

Property	Value
Audited source images	117,546
Final images	109,107
Operational classes	120
Train / validation / test	76,376 / 16,368 / 16,363
Historical duplicate relationships	8,672
Historical cross-split duplicate relationships	3,233

Table 3: Dataset audit and frozen benchmark ledger.

Figure 3: Long-tail training support and pre-training diagnostic probes.

Field	Completed reference configuration
Backbone	facebook/dinov3-convnext-tiny-pretrain-lvd1689m
Input / classes	256 $\times$ 256 / 120
Optimiser	AdamW (betas and epsilon not explicitly preserved)
Backbone / head learning rate	5e-05 / 3e-04
Weight decay	0.05
Epochs / best epoch	40 / 14

Table 4: Completed DINOv3 reference-training configuration.

Panel of Table 5 — Table 5: Validation-only post-training quantisation candidates.

Evidence domain	Archived value
Execution context	Kaggle-host CPU software-runtime evaluation; not Android device evidence
Software stack	Python 3.12.13 ; PyTorch / torchvision 2.12.1 / 0.27.1 ; ExecuTorch / torchao 1.3.1 / 0.17.0 ; timm / FlatBuffers 1.0.26 / 25.12.19
Backend / input	XNNPACK / [1, 3, 256, 256]
Quantisation contract	Requested scope full ; dynamic activation / per-channel weights; explicit module exclusions []
Coverage evidence	$\checkmark$ PASS Lowered backend modules / delegate tokens: 1 / 1
Unarchived host details	CPU model, thread count, and OS build

Table 6: Archived runtime environment and quantisation coverage.

State	Architecture / object	Accuracy (%)	Balanced accuracy (%)	Macro-F1 (%)	Top-3 (%)	Top-5 (%)	Errors
Reference	DINOv3 ConvNeXt-Tiny	$98.5088$	$96.6836$	$96.8700$	$99.7250$	$99.8533$	$244$
Float student	MobileNetV4 Conv-Medium	$98.4599$	$96.2435$	$96.2710$	$99.7494$	$99.8594$	$252$
Converted INT8	XNNPACK-compatible converted graph	$98.4538$	$96.1957$	$96.2492$	$99.7372$	$99.8533$	$253$
PTE runtime	Serialised ExecuTorch/XNNPACK program	$98.4599$	$96.2017$	$96.2267$	$99.7433$	$99.8717$	$252$

Table 7: Performance across identified model states on the locked internal test.

Figure 4: Accuracy, balanced accuracy, and macro-F1 across the four identified model states, with validation-only PTQ candidate results.

Metric	Reference (%)	PTE (%)	PTE $-$ reference (pp)	Paired 95% interval (pp)
Accuracy	$98.509$	$98.460$	$-0.049$	$[-0.214,+0.116]$
Balanced accuracy	$96.684$	$96.202$	$-0.482$	$[-1.097,+0.113]$
Macro-F1	$96.870$	$96.227$	$-0.643$	$[-1.239,-0.047]$
Fruit-category mean recall	$85.936$	$78.971$	$-6.965$	$[-12.409,-1.711]$

Table 8: Paired reference-to-PTE changes with class-stratified percentile intervals.

Training support	Classes	Reference macro-F1 (%)	PTE macro-F1 (%)	Change (pp)	Reference recall (%)	PTE recall (%)	Change (pp)
$<100$	17	91.17	88.60	$-2.56$	89.37	87.27	$-2.10$
100–199	12	92.84	90.05	$-2.78$	93.09	90.27	$-2.82$
200–499	38	97.91	97.85	$-0.06$	98.04	98.34	$+0.30$
500–999	30	99.07	98.98	$-0.09$	99.10	99.01	$-0.09$
$\geq 1000$	23	98.60	98.81	$+0.21$	98.58	98.71	$+0.13$

Table 9: Performance retention by disjoint training-support bin.

Class	Train support	Test support	Reference F1 (0–1)	PTE F1 (0–1)	$\Delta$ F1 (0–1)
banana_disease_fruit	$51$	$11$	$0.7778$	$0.6000$	$-0.1778$
mango_disease_fruit	$143$	$30$	$0.8788$	$0.7213$	$-0.1575$
guava_healthy_fruit	$35$	$7$	$1.0000$	$0.8571$	$-0.1429$
guava_disease_fruit	$110$	$23$	$0.8000$	$0.7234$	$-0.0766$
papaya_fruit_healthy	$96$	$20$	$0.9231$	$0.8500$	$-0.0731$
tomato_mosaic_virus	$240$	$51$	$0.9703$	$0.9231$	$-0.0472$

Table 10: Largest class-level F1 declines.

Figure 5: Reference-versus-PTE class F1; point area is proportional to test support.

Validation target coverage (%)	Reference test coverage (%)	Reference accepted accuracy (%)	PTE test coverage (%)	PTE accepted accuracy (%)
80	79.80	99.824	79.72	99.678
90	89.79	99.816	89.90	99.681
95	94.88	99.813	94.82	99.671
98	97.99	99.451	97.90	99.407

Table 11: Validation-derived selective-prediction operating points.

Figure 6: Validation-derived selective-prediction operating points and the six converted-INT8-to-PTE decision changes.

Appendix figures & tables6 assets

Supplementary material from the paper’s appendix.

Appendix

Object	Identifier
Final retained dataset semantic fingerprint	7c368e6e3d8be3bb3a9a3a5f961075d4 faa125bcac2e98a3b55e1a1c61f1c523
Final dataset build fingerprint	2d7c237981b8943d9b08a522db048984 9a4bfe7f488dfb6461489dd36dcdc12b
Dataset manifest SHA-256	bdb82211ccc2059153724eea178a1680 893a6b38ecc243fae484baa91dbf68e2
Class-map SHA-256	46f7811726c19c42bd7213b2d8178b19 a5a182a1b763f60a94ee2c0e5f6688d2
Reference checkpoint SHA-256	74b4701b8931976c9227845ead50788a e47e3596f575f2817b7352a715f53b79
Mobile configuration fingerprint	a75442087ee1c148b1dcd68441f2fd87 8d8738844532ece4d5606a0038ffb91e

Appendix

Table 12: Model and dataset identity registry.

Metric	Reference point	PTE point	PTE $-$ reference	95% lower	95% upper
Accuracy	0.985088	0.984599	-0.000489	-0.002140	+0.001161
Balanced accuracy	0.966836	0.962017	-0.004819	-0.010967	+0.001128
Macro-F1	0.968700	0.962267	-0.006433	-0.012386	-0.000467
Fruit mean recall	0.859364	0.789711	-0.069653	-0.124086	-0.017108
Recall $<100$	0.893703	0.872686	-0.021017	-0.057207	+0.013036
Recall 100–199	0.930925	0.902687	-0.028238	-0.056711	+0.000623

Appendix

Table 13: Paired bootstrap state-transition summary.

Row ID	True class	Converted INT8	PTE	INT8 confidence	PTE confidence
1455	banana_healthy_fruit	banana_healthy_fruit $\checkmark$	banana_disease_fruit $\times$	0.479	0.477
3891	corn_northern_leaf_blight	corn_northern_leaf_blight $\checkmark$	wheat_leaf_rust $\times$	0.303	0.358
6292	nutrient_deficiency	pear_black_spot $\times$	nutrient_deficiency $\checkmark$	0.486	0.468
11743	sugarcane_yellow_leaf	sugarcane_mosaic $\times$	sugarcane_yellow_leaf $\checkmark$	0.530	0.472
12361	tomato_early_blight	tomato_septoria_leaf_spot $\times$	tomato_early_blight $\checkmark$	0.481	0.477
12750	tomato_healthy	apple_healthy $\times$	cherry_healthy $\times$	0.262	0.267

Appendix

Table 14: Converted-INT8-to-PTE disagreement audit.

Operational error category	Reference errors	PTE errors	Change
Cross-crop condition confusion	52	48	-4
Cross-crop healthy/condition confusion	6	9	+3
Fruit-condition label involved	47	64	+17
Generic plant_healthy involved	22	12	-10
Same-crop condition confusion	88	90	+2
Same-crop healthy/condition confusion	29	29	+0

Appendix

Table 15: Reference-to-PTE operational error taxonomy.

Panel of Table 16 — Table 16: Archived runtime and quantisation environment.

Operational class	Crop/species	Label type	Train	Test
aloevera_healthy	aloevera	Generic healthy state	63	14
amla_healthy	amla	Generic healthy state	49	10
apple_black_rot	apple	Specific named disease	1,426	306
apple_black_spot	apple	Specific named disease	218	47
apple_brown_spot	apple	Specific named disease	696	149
apple_cedar_rust	apple	Specific named disease	202	43

Appendix

Table 17: Operational class taxonomy and split support.

Agricultural Image Analysis

Agricultural image analysis uses computer vision and multimodal learning to monitor crops, livestock, produce quality, and field conditions.

See all 94 papers

A100: A Hybrid Vision Foundation Model for Agricultural Plant Image Analysis
Oct 16, 2026Mayur Pratim Das, Ajai Dubey, Aksshay Mathew P +5Agricultural Image Analysis Vision Foundation Models
Annam.AI, IIT Ropar Rupnagar, India · IIT Ropar Rupnagar, India · Dept. of DSE, IISER Bhopal Bhopal, India +1
SAGE: Scalable Agentic Grounded Evaluation for Crop Disease Diagnosis
May 10, 2026Muhammad Arbab Arshad, Tirtho Roy, Yanben Shen +7VLM Evaluation Agricultural Image Analysis
Iowa State University, USA · New York University, USA
AgriPath: A Systematic Exploration of Architectural Trade-offs for Crop Disease Classification
Mar 8, 2026Hamza Mooraj, George Pantazopoulos, Alessandro SugliaContrastive Learning for VLMs Plant Disease Classification
School of Mathematical and Computer Sciences Heriot-Watt University · Institute for Language, Cognition and Computation, School of Informatics University of Edinburgh

Oct 16, 2026J/K move · Enter open · S save

A100: A Hybrid Vision Foundation Model for Agricultural Plant Image Analysis

Mayur Pratim Das, Ajai Dubey, Aksshay Mathew P, Sana Afreen+4

Annam.AI, IIT Ropar Rupnagar, India · IIT Ropar Rupnagar, India · Dept. of DSE, IISER Bhopal Bhopal, India+1

Agricultural Image Analysis Vision Foundation Models Plant Disease Classification Plant Phenotyping

Open paper AskPDF arXiv

The use of accurate vision models for crop monitoring, disease diagnosis, and plant phenotyping in agriculture is hindered by the lack of expertly annotated data, especially in developing countries. We propose A100, an agricultural vision foundation model that is pre-trained using self-supervised learning on 3 million unlabeled images from fields covering more than 100 crop species. A100 uses a hybrid CNN-Transformer backbone that fuses MBConv stages for local feature extraction with Transformer stages for global contextual representation. A100 uses pre-training with the integration of DINO self-distillation, iBOT masked patch reconstruction, and KoLeo uniformity regularization, resulting in features that are better than ImageNet-12K supervised representations for agricultural applications without any fine-tuning. By utilizing a frozen backbone, the accuracy for crop classification of A100 is 95.37%, which is +2.56% better than ImageNet-12K, and reduces the leaf count error by 10.4% on the GroMo25 phenotyping benchmark, proving that the same backbone can be used for classification and regression tasks. After full fine-tuning, the performance of A100 is 99.90% on 32-class crop identification and 99.02% on 44-class disease detection. On the difficult PlantDoc dataset, A100 outperforms all baselines, such as CvT-13 (64.89% against 63.56%). Remarkably, A100 obtains 98.01% accuracy on plant classification using just 1% of the labelled data (about 284 images per category), showing great label efficiency for use cases with limited labeling. Under an identical protocol, A100 also outperforms the publicly released agricultural foundation models FoMo4Wheat and LeafVision by more than 14 points at 1% labels. We plan to publicly release our pretrained model to allow agricultural professionals to adapt A100 for various plant analysis applications using limited annotations.

Abstract

Agricultural Image Analysis

Explore similar work

A100: A Hybrid Vision Foundation Model for Agricultural Plant Image Analysis

SAGE: Scalable Agentic Grounded Evaluation for Crop Disease Diagnosis

AgriPath: A Systematic Exploration of Architectural Trade-offs for Crop Disease Classification