Metraipolis › archive › dc_db381599a4fb
TinyCNN: A 193K-Parameter Network for On-Device Plant Disease Detection, with a Cross-Dataset Robustness Diagnosis — arXiv:2609.20290
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.20290
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-10T04:09:22.308Z
- SHA-256 of the archived text
c49ef155085d0f1206e73dc3c3f8e5cc66bdbe289760d9973742fecc593ea188- Stable URL of this page
https://metraipolis.com/documento/dc_db381599a4fb
How to cite
TinyCNN: A 193K-Parameter Network for On-Device Plant Disease Detection, with a Cross-Dataset Robustness Diagnosis — arXiv:2609.20290 — arXiv (en), https://arxiv.org/abs/2609.20290 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_db381599a4fb (Metraipolis archive dc_db381599a4fb, sha256 c49ef155085d0f1206e73dc3c3f8e5cc66bdbe289760d9973742fecc593ea188)
BibTeX
@misc{dcdb381599a4fb,
title = {TinyCNN: A 193K-Parameter Network for On-Device Plant Disease Detection, with a Cross-Dataset Robustness Diagnosis — arXiv:2609.20290},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.20290},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_db381599a4fb; sha256 c49ef155085d0f1206e73dc3c3f8e5cc66bdbe289760d9973742fecc593ea188},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.20290. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Detecting crop disease early is central to sustainable agriculture and food security under United Nations Sustainable Development Goal 2 (Zero Hunger), and is especially urgent in resource-constrained regions where expert diagnosis is scarce but low-cost mobile devices are widespread. This paper presents TinyCNN, a lightweight convolutional neural network for on-device plant disease classification. TinyCNN uses depthwise separable convolution blocks and contains only 193,190 trainable parameters with 110.05M MACs for a 224x224 input image. On the 38-class PlantVillage benchmark, TinyCNN achieves 98.88% test accuracy and 98.03% macro-F1 while being approximately 58x smaller than ResNet18 and 11.8x smaller than a MobileNetV2 teacher, directly reducing the energy, memory, and cost footprint of inference in line with Green AI principles. The paper further analyzes vanilla knowledge distillation as a sustainable model-compression strategy; an ablation over alpha in {0.3, 0.5, 0.7} and T in {2, 4} selects alpha=0.3, T=4, producing a distilled TinyCNN with 98.81% test accuracy. Finally, cross-dataset evaluation from PlantVillage to PlantDoc reveals a substantial robustness gap under real-world conditions, which a Grad-CAM analysis attributes to off-leaf, background-driven attention consistent with shortcut learning. TinyCNN is thus an energy-efficient, deployable building block for sustainable agricultural intelligence, while field robustness remains the key barrier to durable real-world impact.
Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-18 in cs.LG).
Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV) ; Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.20290 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-10T03:48:18.440Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).