Mango_deep_yield_dataset.zip (137.04 MB)
Mango_deep_yield_dataset koirala et al. 2021
dataset
posted on 2021-02-15, 05:46 authored by Anand KoiralaAnand Koirala, Zhenglin WangZhenglin Wang, Kerry WalshKerry WalshKoirala, A.; Walsh, K.B.; Wang, Z. Attempting to Estimate the Unseen – Correction for Occluded Fruit in Tree Fruit Load Estimation by Machine Vision with Deep Learning.
This dataset contains dual-view images (image from two opposite sides (sideA and sideB) of a tree) used in the paper "Attempting to estimate the unseen - correction for occluded fruit in tree fruit load estimation by machine vision with deep learning".
There are three orchards A, B and C with images of same trees from two seasons (2017 and 2018).
For each season ABC is the collection of images from orchards A, B and C put together.
A-x, B-x and C-x comprise of extended set of images collected in season 2017.
A= 17 trees
B= 6 trees
C= 12 trees
ABC= 35 trees
A-x= 44 trees
B-x= 19 trees
C-x= 35 trees
ABC-x = 98 trees
harvest_data_deep_yield.xlsx file contains the harvest fruit count mapped with the tree and image names for each orchard.
Funding
Project ST19009, Multiscale monitoring tools for managing Australian tree crops.
History
Peer Reviewed
- Yes
Open Access
- Yes
Acceptance Date
2021-02-10Author Research Institute
- Institute for Future Farming Systems
Era Eligible
- Yes