Dedicated dataset of Sapota (Manilkara zapota) fruit for machine vision applicationsMendeley Data and Zenodo
Machine vision and robotics can play essential roles in solving agricultural problems. The collection and creation of datasets is an indispensable step for machine vision-based applications. Sapota (Manilkara zapota) is an evergreen fruit grown in many regions in India. Fruit samples are collected f...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2025-08-01
|
| Series: | Data in Brief |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2352340925006201 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | Machine vision and robotics can play essential roles in solving agricultural problems. The collection and creation of datasets is an indispensable step for machine vision-based applications. Sapota (Manilkara zapota) is an evergreen fruit grown in many regions in India. Fruit samples are collected from Gujarat state, and three datasets of the Kalipatti variety of sapota fruit images are created. An original dataset was created using images of fresh and spoiled sapota fruits based on their physical and visual characteristics with white background under different lighting conditions. Fresh fruits are further classified into small, medium, and large categories. Two versions of this dataset exist: Dataset-1, which consists of resized images with the white background retained, and Dataset-2, which presents the images in their original dimensions with the background removed and includes corresponding annotation files. Dataset-3 was created using images of fresh and spoiled sapota fruit against various backgrounds, such as grass, soil, blue, and white. Our dataset contains 16,826 sapota fruit images in diverse backgrounds and lighting conditions. These datasets will help create machine vision-based real-time sorting and grading systems to increase the effective yield of sapota fruit. |
|---|---|
| ISSN: | 2352-3409 |