The dataset consists of multiple repositories:
-
Fruits-360 100x100 (Images scaled to 100x100 pixels. Currently contains 166293 pictures of 237 fruits, vegetables, nuts and seeds.)
-
Fruits-360 original-size (Original (captured) size images. Currently contains 86899 pictures of 122 fruits, vegetables, nuts and seeds.)
-
Fruits-360 3-body-problem (A 3-class classification problem, where the Training and Test folders contain different (varieties of) the 3 fruits and vegetables (Apples, Cherries and Tomatoes). Currently contains 47033 pictures of Apples, Cherries and Tomatoes.)
-
Fruits-360 meta (Attributes. Currently contains data of 26 fruits and vegetables.)
-
Fruits-360 multi (Multiple objects in the same picture. Currently contains 150 images.)
-
Fruits-360 papers (Papers using Fruits-360 dataset.)
email: [email protected]
Mihai Oltean, Fruits-360 dataset, 2017-.
-
2016 (begining of). I met with a private company that needed a robot to handle the goods from their food stores. My solution was to use Jenny 5 robot for this purpose. Even if Jenny 5 robot has three Logitech C910 cameras, we had no serious algorithm capable of recognizing food. So, the first step was to build a dataset with many images of foods that could be fed into a neural network as training data ...
-
2016 (at the end). I started collecting fruits (either from my own garden or purchased from local stores). I stuck them into a low-rpm motor and filmed them. Initially, the films were made inside my room, which has blue-painted walls, whose reflection is visible in many images from that time.
-
2017 (begining of). A student of mine wrote an example of using TensorFlow for the Fruits-360 dataset.
-
2017 (summer of). The student presented his work at a local symposium for students.
-
2017 (end of). A more complete paper with the dataset and numerical experiments has been accepted and publised by Acta Univ. Sapientiae. The paper has been updated several times after publication to include results for newer dataset versions.
-
2017 (end of). The dataset has been uploaded to Kaggle: https://www.kaggle.com/datasets/moltean/fruits.
-
2021 (begining of). Original (captured) size images will be published from now on.
-
2021 (begining of). The metadata (properties) of fruits will be included in the dataset.
-
2024.02.17. The dataset has been moved to a new location on GitHub: https://gihub.com/fruits-360. The organization contains multiple repositories.
-
2025.03.23. New repository with papers using Fruits-360 dataset: https://github.com/fruits-360/fruits-360-papers
-
2025.04.01. New dataset: Fruits-360, 3-body problem: https://github.com/fruits-360/fruits-360-3-body-problem
-
Mihai Oltean, Fruits-360 dataset: new research directions, Technical report, 2021.
-
Horea Muresan, Mihai Oltean, Fruit recognition from images using deep learning, Acta Univ. Sapientiae, Informatica, Vol. 10, Issue 1, pp. 26-42, 2018.
More papers using Fruits-360 dataset are here: https://github.com/fruits-360/fruits-360-papers
All fruits, vegetables, nuts, seeds etc, are from:
- my own garden (located in Cugir, Romania) or
- purchased, by me, from local stores.
I have recorded all the movies containing the fruits, vegetables, etc in the Fruits-360 dataset.
I have extracted (from movies) all pictures in the Fruits-360 dataset.
I have NOT taken pictures from other sources.
I have NOT generated artificial pictures of fruits, vegetables, nuts or seeds.
CC BY-SA 4.0
Copyright (c) 2017-, Mihai Oltean
You are free to:
Share — copy and redistribute the material in any medium or format for any purpose, even commercially.
Adapt — remix, transform, and build upon the material for any purpose, even commercially.
Under the following terms:
Attribution — You must give appropriate credit , provide a link to the license, and indicate if changes were made . You may do so in any reasonable manner, but not in any way that suggests the licensor endorses you or your use.
ShareAlike — If you remix, transform, or build upon the material, you must distribute your contributions under the same license as the original.