pratheesh3780/banana_images_cnn

Banana Image Dataset

Banana image dataset comprising 4320 banana images belongs to 4 ripening classes namely unripe, medium ripe, ripe and overripe were collected as part of Master's project in Agricultural Statistics on the topic ‘An intelligent ripening classification of nendran banana (Musa spp.) using Convolutional Neural Network (CNN).The images were captured using an Android device equipped with a 50 MP camera, ensuring detailed visual information for each sample. To maintain consistency and eliminate potential confounding factors, the banana samples were placed on a table covered with non-reflective white paper, providing a uniform background across all images. The data collection process was conducted under white light illumination, enhancing the accuracy and reliability of the dataset. The unripe, medium ripe , ripe and overripe classes contain 1155,1060,992 and 1113 images respectively.This image dataset is made available as open source under the CC BY License 4.0, allowing for free download and utilization in the development of classification models or the enhancement of existing ones. The license, with its "BY" element, mandates proper credit to the creator, while also granting others the freedom to redistribute, remix, transform, and utilize the material for various purposes, including commercial applications. This open-source dataset holds significant potential for advancing the field of intelligent ripening classification, facilitating research endeavors and the improvement of existing models.