Skip to content

Fruit detection dataset

This dataset is relatively small (0.72GB) so that it should be fine with any option. We provide 4 download options; via Zenodo, Kaggle API, and wget.

Note

In order to access this, you will need to create Roboflow account using your work email or github account (free of charge).

Roboflow is one of convenient image annotation, and model training platforms. It allows sharing annotations with others and collaborative cloud-based annotations. Please have look if you are interested in this and more than welcome to contribute our incremental dataset 😄. You can find all links from table below.

Dataset # instances Roboflow link
🍎 Apple 354 roboflow apple
🥑 Avocado 178 roboflow avocado
🫐 Blueberry 3176 roboflow blueberry
🫑 Capsicum 724 roboflow capsicum
🍒 Cherry 4137 roboflow cherry
🥝 Kiwi 3716 roboflow kiwi
🥭 Mango 1145 roboflow mango
🟠 Orange 274 roboflow orange
🍈 Rockmelon 187 roboflow rockmelon
🍓 Strawberry 341 roboflow strawberry
Wheat 147793 roboflow wheat

Tip

You can export these annotations in 17 different formats by simply selecting Export from Roboflow.

Note

In order to use this option, you will need to setup Kaggle API on your computer. Please have a look Kaggle API for detail.

This option is the most reliable based on our experience for large files transaction but it requires to setup your Kaggle API. Please make sure the API is operational by testing a couple of examples from Kaggle API.

kaggle datasets download -d enddl22/deepnir-11fruits
Alternatively, you can also download from link below

Kaggle 11 fruits

We also host our dataset using Zenodo, one of general-purpose open-access repositories. You can find them from the following links

Dataset Description Size (GB)
deepNIR fruits 11 fruits and their bounding box 0.72

Warning

You may encounter ERROR 403: Forbidden. or Sorry, you can't view or download this file at this time. Too many users have viewed or downloaded this file recently.... errors. This happens due to limited acceess to gdrive and should work within 24 hours but not guranteed. It is thus recommended to use other options in case you download larger files.

You can also directly download from our gdrive with the command below. It is a bit complicated command due to cookie business but if you want to use this option, simply copy the command line (either dragging or click Copy to clipboard botton when your cursor hover the command.) and paste it to your terminal console. Download speed may vary depending on your Internet connection.

11 Fruits images and their annotation (YOLOv5 format)
wget --load-cookies /tmp/cookies.txt "https://docs.google.com/uc?export=download&confirm=$(wget --quiet --save-cookies /tmp/cookies.txt --keep-session-cookies --no-check-certificate 'https://docs.google.com/uc?export=download&id=16Q50eQBL9ZLWh6a4jKI390HmogzfhDuQ' -O- | sed -rn 's/.*confirm=([0-9A-Za-z_]+).*/\1\n/p')&id=16Q50eQBL9ZLWh6a4jKI390HmogzfhDuQ" -O yolov5.zip && rm -rf /tmp/cookies.txt

How can I uncompress download files?

We compressed all files in this page as zip format under Ubuntu system. You can use the following command to uncompress them. More detail regarding how to uncompress files can be found from here

unzip yolov5.zip
unzip yolov5.zip

Just right click on the file and select uncompress. Alternatively, you can install 7-zip on your Windows system to uncompress the file.