Research Article
Early Ginger Disease Detection Using Deep Learning Approach
@INPROCEEDINGS{10.1007/978-3-030-93709-6_32, author={Mesay Gemeda Yigezu and Michael Melese Woldeyohannis and Atnafu Lambebo Tonja}, title={Early Ginger Disease Detection Using Deep Learning Approach}, proceedings={Advances of Science and Technology. 9th EAI International Conference, ICAST 2021, Hybrid Event, Bahir Dar, Ethiopia, August 27--29, 2021, Proceedings, Part I}, proceedings_a={ICAST}, year={2022}, month={1}, keywords={CNN Deep learning Image detection Image classification Ginger disease}, doi={10.1007/978-3-030-93709-6_32} }- Mesay Gemeda Yigezu
Michael Melese Woldeyohannis
Atnafu Lambebo Tonja
Year: 2022
Early Ginger Disease Detection Using Deep Learning Approach
ICAST
Springer
DOI: 10.1007/978-3-030-93709-6_32
Abstract
Ethiopia is one of the countries in Africa which have a huge potential for the development of different varieties of crops used for traditional medicine and daily use in society. Ginger is one among the others which are affected by disease caused by bacteria, fungi, and virus being bacterial wilt is the most determinant constraint to ginger production. Detection of the disease needs special attention from experts which is not possible for mass production. However, the state-of-the-art technology can deploy to overcome the problem by means of image processing in a mass cultivates ginger crop. To this end, a deep learning approach for early ginger disease detection from the leaf is proposed through different phases after collecting 7,014 ginger images with the help of domain experts from different farms. The collected data passed through different image pre-processing to design and develop a deep learning model that can detect and classify with a different scenario. The experimental result demonstrates that the proposed technique is effective for ginger disease detection especially bacterial wilt. The proposed model can successfully detect the given image with a test accuracy of 95.2%. The result shows that the deep learning approach offers a fast, affordable, and easily deployable strategy for ginger disease detection that makes the model a useful early disease detection tool and this analysis is also extended to develop a mobile app to help a lot of ginger farmers in developing countries.
