EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Hybrid Lightweight Deep Learning Architecture for Small-Scale Millet Classification

Baisakh

Journal of Intelligent Decision Making and Information Science · 27.08.2026 · 3 (9s) · 1267-1288

DOI: 10.17762/jidmis.v3.2485 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Journal of Intelligent Decision Making and Information Science, online erschienen am 27.08.2026. Verfasst von Baisakh. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

In recent time, millets have emerged as one of the most sustainable solutions to global issues like food insecurity, climate change; however, the incorporation of Artificial Intelligence in precision farming frameworks, remains significantly unexplored. In this article, a novel combination of both supervised and unsupervised model is proposed for a small scale fine grain millet classification. This framework integrates the light weight encoder with a pretrained VGG-19 network architecture which produces efficient and accurate classification of distinct millet seed varieties. To enable fair evaluation of the proposed model under the resource constraint conditions, a priority based performance metrics is used while comparing with different deep neural network models. The priority weight is calculated based on the inference time, total trainable parameters and classification accuracy. The proposed auto encoder based hybrid VGG19 model achieved the highest overall performance score 0.960 with 93.60% classification accuracy and an inference time of 0.4418 seconds, outperforming InceptionV3, DenseNet121, ResNet50, MobileNetV2, and AE-only baselines.

Zitieren

APA

Baisakh (2026). Hybrid Lightweight Deep Learning Architecture for Small-Scale Millet Classification. Journal of Intelligent Decision Making and Information Science, 3(9s), 1267-1288. https://doi.org/10.17762/jidmis.v3.2485

Vancouver

Baisakh. Hybrid Lightweight Deep Learning Architecture for Small-Scale Millet Classification. Journal of Intelligent Decision Making and Information Science. 2026;3(9s):1267-1288. doi:10.17762/jidmis.v3.2485

BibTeX
@article{Baisakh2026,
  author = {Baisakh},
  title = {Hybrid Lightweight Deep Learning Architecture for Small-Scale Millet Classification},
  journal = {Journal of Intelligent Decision Making and Information Science},
  year = {2026},
  volume = {3},
  number = {9s},
  pages = {1267--1288},
  doi = {10.17762/jidmis.v3.2485}
}
Kostenlos starten →