Datasets:
license: apache-2.0
task_categories:
- image-classification
language:
- en
tags:
- multiclass-image-classification
size_categories:
- n<1K
Fruits30 Dataset
Description:
The Fruits30 dataset is a collection of images featuring 30 different types of fruits. Each image has been preprocessed and standardized to a size of 224x224 pixels, ensuring uniformity in the dataset.
Dataset Composition:
- Number of Classes: 30
- Image Resolution: 224x224 pixels
- Total Images: 856
Classes:
"0" : "acerolas"
"1" : "apples"
"2" : "apricots"
"3" : "avocados"
"4" : "bananas"
"5" : "blackberries",
"6" : "blueberries",
"7" : "cantaloupes",
"8" : "cherries",
"9" : "coconuts",
"10" : "figs",
"11" : "grapefruits",
"12" : "grapes",
"13" : "guava",
"14" : "kiwifruit",
"15" : "lemons",
"16" : "limes",
"17" : "mangos",
"18" : "olives",
"19" : "oranges",
"20" : "passionfruit",
"21" : "peaches",
"22" : "pears",
"23" : "pineapples",
"24" : "plums",
"25" : "pomegranates",
"26" : "raspberries",
"27" : "strawberries",
"28" : "tomatoes",
"29" : "watermelons"
Preprocessing:
Images have undergone preprocessing to maintain consistency and facilitate model training. Preprocessing steps may include resizing, normalization, and other enhancements.
Intended Use:
The Fruits30 dataset is suitable for tasks such as image classification, object recognition, and machine learning model training within the domain of fruit identification.
Sources:
Croudsource.
Note:
Ensure proper attribution and compliance with the dataset's licensing terms when using it for research or development purposes.