Skip to content
Open access

Saturated In-Domain, Separable Out-of-Domain: A Hierarchical Multi-Scale Lesion-Attention Network and a Zero-Shot Cross-Corpus Protocol for Multi-Crop Leaf Disease Classification

Sep 2026 · Symmetry · Vol 18, pp. 1557 · 0 citations · 35 references

TL;DR

This work proposes HMLA-Net, which adds multi-scale fusion, gated multi-head lesion-aware pooling, an auxiliary crop head and MixStyle to a CAFormer-S18 backbone and averages the posterior over the Klein four-group of flips at inference.

Abstract

Leaf disease classifiers are ranked by held-out accuracy on their training corpus, where that number now exceeds 98% for many modern backbones. We ask what such a measurement can still resolve. On a four-crop, 21-class corpus of 7179 de-duplicated images, we train 26 architectures under one protocol with three seeds and shared partitions, and we execute the entire benchmark twice. The 26 architectures span 1.68 percentage points (pp) of internal macro-F1. The same configuration differs by 0.34 pp on average and by up to 1.06 pp between the two executions, and the internal ordering only partly replicates (Spearman’s ρ=0.69, with VGG-19 moving from 17th to 1st). Applied zero-shot to an independently collected mango corpus, the same checkpoints spread over 47 pp of accuracy and their ordering does replicate (ρ=0.92). We propose HMLA-Net, which adds multi-scale fusion, gated multi-head lesion-aware pooling, an auxiliary crop head and MixStyle to a CAFormer-S18 backbone and averages the posterior over the Klein four-group of flips at inference. In the reference execution, it attains the highest internal macro-F1 (0.9845) and the highest zero-shot accuracy (0.699) of the 26 architectures, with the smallest degradation (28.9 pp) and the lowest rate of predictions leaking into labels absent from the target corpus. The second execution reproduces its zero-shot rank but not its internal one. Five backbones retrained with its complete recipe show that its margin over its own backbone lies within replicate variation, and in a 16-removal ablation measured on both partitions, only the large effect of ImageNet pretraining remains consistent across the two executions. A cross-corpus duplicate audit finds no exact duplicates and no evidence of substantial overlap, and one class, mango sooty mould, collapses to near-zero recall in 19 of 26 architectures. The contribution is the protocol: replicate the benchmark, evaluate out of domain, and report results per class.

Read PDF

Similar papers

Open access Oct 2026

MSAF-Net: Multi-Scale Attention Fusion for Fine-Grained Medicinal Plant Recognition Under Near-Duplicate-Audited and Cross-Dataset Evaluation

Recognising medicinal plant species from leaf images is a fine-grained problem: silhouette and venation provide complementary evidence at different feature scales. We propose MSAF-Net, which taps three stages of an ImageNet-22k pre-trained ConvNeXt-Tiny backbone, refines each stage with convolutional block attention, l...

Davut Ari, Mehmet Burukanli, Songul Karakus · 0 citations
Open access Sep 2026

Beyond Single-Corpus Leaderboards: Controlled Cross-Corpus Evaluation of Frequency-Gated and Multi-Scale Cross-Attention Networks for Image Deepfake Detection

Image deepfake detectors are routinely selected on single-corpus accuracy, yet deployment requires a decision on media whose manipulation family is unknown. This paper asks whether single-corpus ranking predicts ranking under unknown provenance, and answers in the negative. Six architectures are trained from a common i...

Abhay Mundra, M. Meena · 0 citations
Preprint Oct 2026

Multi-Crop Leaf Disease Recognition: A Unified Benchmark and Cross-Region Study

Deep learning models for crop leaf disease recognition routinely report near-perfect accuracy yet are typically trained and evaluated on a single dataset collected under controlled laboratory conditions, leaving their behavior under realistic cross-region domain shift poorly understood. We introduce MLD (Multi-crop Lea...

R. Nalwanga, Sebastian Bunda, G. Owomugisha et al. · 0 citations
Open access Aug 2026

A Controlled Evaluation of Dual-Channel Feature Enhancement and Multi-Level Knowledge Distillation for Lightweight Plant Disease Recognition

DC-FEN, a MobileNetV3-based design that models spatial-token relations and channel interactions in parallel and injects them through gated residual fusion is introduced and shows that adding intermediate transfer constraints does not guarantee a stronger student.

Xin Lei, Yong-Huai Liu, Ardhendu Behera et al. · 0 citations
Preprint Sep 2026

Cross-Dataset Generalization of Bangladeshi Rice Leaf Disease Classifiers: Benchmark, Diagnosis, and Mitigation

Cross-dataset transfer in rice leaf disease classification remains a significant challenge, with models trained on one image collection performing substantially worse when deployed on another. We conduct a systematic benchmark across three Bangladeshi rice leaf disease datasets (5,419 images, 6 transfer pairs, 3 CNN ba...

Anindya Paul · 0 citations

We use cookies to run the site and, with your consent, for analytics and to show ads. See our Cookie Policy.