News & Updates

• (2025-03-20) — New Preprint on arxiv: MobilePlantViT: A Mobile-friendly Hybrid ViT for Generalized Plant Disease Image Classification - [Paper]

  We introduce MobilePlantViT, a lightweight Vision Transformer for plant disease classification, achieving 80%-99% accuracy with just 0.69 million parameters, ideal for mobile and edge device deployment in smart agriculture.

• (2025-03-19) — Paper published in Agronomoy Journal: Attention Mechanism-based Ultralightweight Deep Learning Method for Automated Multi-fruit Disease Recognition System - [Paper]

  We propose an ultralightweight deep learning architecture for fruit disease recognition, achieving over 99% accuracy on sugar apple, pomegranate, and guava datasets. Our model integrates advanced techniques to ensure high performance and efficiency, making it ideal for memory-constrained edge devices in smart agriculture.