MinneApple: A Benchmark Dataset for Apple Detection and Segmentation

Loading...
Thumbnail Image
Statistics
View Statistics

Collection period

2015-06-01
2016-09-30

Date completed

2019-09-07

Date updated

Time period coverage

Geographic coverage

Source information

Journal Title

Journal ISSN

Volume Title

Title

MinneApple: A Benchmark Dataset for Apple Detection and Segmentation

Published Date

2019-09-11

Author Contact

Haeni, Nicolai
haeni001@umn.edu

Type

Dataset
Experimental Data
Field Study Data

Abstract

We present a new dataset with the goal of advancing the state-of-the-art in fruit detection, segmentation, and counting in orchard environments. We hope to achieve this by providing a large variety of high-resolution images acquired in orchards, together with human annotations of the fruits on the trees. Objects are labeled using polygon masks for each object instance to aid in precise object detection, localization or segmentation. Additionally, we provide data for patch-based counting of clustered fruits. Our dataset contains over 40'000 annotated object instances in 1000 images.

Description

The counting and detection data are used to solve two different subproblems of the larger problem that is yield estimation. Yield estimation relies on accurate detection of the fruit. Since fruit can be clustered together, it is necessary to use a separate algorithm for counting (in most cases). We provide data to develop and test algorithms for both subproblems.

Referenced by

Häni, N, Roy, P, Isler, V. A comparative study of fruit detection and counting methods for yield mapping in apple orchards. Journal of Field Robotics. 2019; 1– 20.
https://doi.org/10.1002/rob.21902
N. Häni, P. Roy, and V. Isler, “MinneApple: A Benchmark Dataset for Apple Detection and Segmentation,” IEEE Robotics and Automation Letters, pp. 1–1, 2020,
https://doi.org/10.1109/LRA.2020.2965061

Related to

Replaces

Publisher

Funding information

USDA NIFA MIN-98-G02

item.page.sponsorshipfunderid

item.page.sponsorshipfundingagency

item.page.sponsorshipgrant

Previously Published Citation

Suggested citation

Haeni, Nicolai; Roy, Pravakar; Isler, Volkan. (2019). MinneApple: A Benchmark Dataset for Apple Detection and Segmentation. Retrieved from the University Digital Conservancy, https://doi.org/10.13020/8ecp-3r13.
View/Download file
File View/OpenDescriptionSize
MinneApple Data README.txtDescription of data5.17 KB
counting.tar.gzPatch-based fruit counting dataset (2,875 JPG files and 67,990 PNG files)976.89 MB
detection.tar.gzFruit detection and segmentation dataset (1,671 PNG files)1.7 GB
test_data.zipContains test labels for counting, detection, and segmentation. Made available on 2022-09-121.17 GB

Content distributed via the University Digital Conservancy may be subject to additional license and use restrictions applied by the depositor. By using these files, users agree to the Terms of Use. Materials in the UDC may contain content that is disturbing and/or harmful. For more information, please see our statement on harmful content in digital repositories.