Trained Model Weights and Supporting Data for "A Two-Stage, Semi-Supervised Deep Learning Framework for the Detection and Classification of Ambient Pollen using Evanescent Wave Scattering"

This dataset contains the trained PyTorch model weights (best.pt) and supporting images associated with the manuscript entitled "A Two-Stage, Semi-Supervised Deep Learning Framework for the Detection and Classification of Ambient Pollen using Evanescent Wave Scattering," submitted to Environmental Science & Technology (Manuscript ID: es-2025-16844x). The model was developed using Evanescent Wave Scattering Microscopy (EWSM) data and a semi-supervised deep learning pipeline for ambient pollen detection and classification into 19 morphologically coherent taxonomic groups.