Skip to content
Publications

Dataset · Zenodo · March 30, 2026

CPD-1 Train/Val/Test Splits and Augmented Data for Pollen Grain Classification

Abdinabi Mukhamadiyev, Rustam Safarov, Jumanov Isroil Ibragimovich, Sunatillo Kholmonov, Toshtemir Abdikhafizovich Khujakulov, Doston Khasanov

Abstract

This dataset contains the training/validation/test partitions (Dataset 1) and an alternative train/test partition (Dataset 2), together with the augmented training set, used in the study 'Comparative Analysis of Methods for Improving the Classification Accuracy of Morphologically Similar Pollen Grain Classes on the CPD-1 Dataset'. It is derived from the Cretan Pollen Dataset v1 (CPD-1), originally created by Tsiknakis, N.; Savvidaki, E.; Kafetzopoulos, S.; Manikis, G.; Vidakis, N.; Marias, K.; Alissandrakis, E., and openly available at https://doi.org/10.5281/zenodo.4756361 under CC BY 4.0. Please cite both the original CPD-1 dataset and this derivative when using these splits.

Smart Agriculture and AIAdvanced Neural Network ApplicationsRemote-Sensing Image Classificationpollen grain classificationdata augmentationplant taxonomy

More by Doston Khasanov