5 files

Segmentation of sandplain lupin weeds from morphologically similar narrow-leafed lupins in the field

posted on 2023-03-26, 05:32 authored by Monica DanileviczMonica Danilevicz, Roberto Lujan Rocha, Jacqueline Batley, Philipp E. Bayer, Mohammed Bennamoun, David Edwards, Michael Ashworth

Dataset for weed detection used in the paper "Segmentation of sandplain lupin weeds from morphologically similar narrow-leafed lupins in the field".

The scripts developed are available on GitHub at mdanilevicz/WeedDetectionML. 

This repository contains five datasets collected in the field by drone or smartphone in Western Australia, Australia. The images for field-1, field-2, grow-1 and grow-2 datasets were collected using the DJI Phantom 4 unoccupied aerial vehicle (UAV) RGB camera. The images were collected between 12 and 2 PM under overcast or clear sky conditions. The details for the image collection can be seen in Table 3. The images were collected with 75% side overlap and 80% front overlap, with five ground control points distributed across the field to increase GPS accuracy. Additionally, 217 images with 4879 sandplain lupins labelled among narrow-leafed lupins were downloaded from the Weed-AI database and named ext-1 dataset.

In this repository you will find the RGB images, plant-soil segmentation mask obtained using CIVE vegetation index and weed-crop-soil segmentation mask.

Detailed methods fro dataset generation is available on the paper, and github repositories.


The authors would like to thank the Pawsey Supercomputing Centre for computation resources. The Australian Government supported this work through the Australian Research Council (Projects DP210100296, DP200100762, and DE210100398) and the Grains Research and Development Corporation (Projects 9177539 and 9177591). Monica F. Danilevicz was supported by the Research Training Program scholarship and the Forrest Research Foundation.