Dry Fish Image Dataset

Published: 15 December 2025| Version 1 | DOI: 10.17632/f3sr9xr8md.1
Contributors:
Md Jakir Hossain, Amran Hossain, Iffat Ara Arin, MD Jisan Mia, Abir Hasan

Description

The collected dataset consists of twelve classes of dried fish, each representing a distinct species commonly found in local markets in Dhaka, Bangladesh. Multiple samples from each class were purchased and photographed under consistent conditions, following the procedure described in the Dataset Collection section. In total, 3,533 images were acquired, providing a diverse set of samples for model training and evaluation. The dataset captures variations in shape, size, texture, and color of the dried fish, ensuring that models can learn to differentiate between classes effectively.

Files

Institutions

  • East West University

Categories

Food Science, Fish, Image Acquisition, Deep Learning

Licence