Soybean Seeds

Published: 30 May 2023| Version 6 | DOI: 10.17632/v6vzvfszj6.6
Contributors:
Wei Lin, Youhao Fu,
,
, Daoyi Ma,
,
,
, Qin Su

Description

The dataset includes five-type of individual soybean seed images, totaling 5513: intact, spotted, immature, broken, and skin-damaged. Furthermore, there are over 1000 images of soybean seeds in each category. Those images of individual soybeans were classified into five categories based on the Standard of Soybean Classification (GB1352-2009). The soybean images with the seeds in physical touch were captured by an industrial camera. Subsequently, individual soybean images (227×227 pixels) were divided from the soybean images (3072×2048 pixels) using an image-processing algorithm with a segmentation accuracy of over 98%. The dataset can serve to study the classification or quality assessment of soybean seeds.

Files

Institutions

Jiangsu University of Science and Technology, Nanjing Agricultural University

Categories

Agricultural Science, Computer Science, Soybean

Licence