Introduces USU-Corn-WeedDB, a UAV RGB dataset with 800 annotated 640x640 patches containing 10,539 instances of three weed species in forage corn, benchmarked with 28 object detection models achieving mAP@0.5 of 0.773-0.840.
A comprehensive survey of image augmentation techniques for deep learning,
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2representative citing papers
Empirical comparison of Outlierness, Diversity, Representativeness, Uncertainty, and Random selection for trajectory data augmentation across four datasets shows conditional gains in stability over random baselines but degradation in dense data.
citing papers explorer
-
USU-Corn-WeedDB: A UAV RGB Image Dataset for Multi-Species Weed Detection in Forage Corn
Introduces USU-Corn-WeedDB, a UAV RGB dataset with 800 annotated 640x640 patches containing 10,539 instances of three weed species in forage corn, benchmarked with 28 object detection models achieving mAP@0.5 of 0.773-0.840.