An automated approach for eggplant disease recognition using transfer learning

Izazul Haque Saad, Md. Mazharul Islam, Isa Khan Himel, Md. Jueal Mia

Abstract


In Bangladesh, eggplant is a widely grown crop that is vital to the country’s food security. The vegetable is consumed on a regular basis by the majority of people. Since Bangladesh’s economy is heavily reliant on agriculture, eggplant growing might help the country’s economy and productivity flourish more efficiently. It provides several health benefits, including reducing cancer risk, blood sugar control, heart health, eye health, and others. Although eggplant is a valuable crop, it is subject to severe diseases that reduce its productivity. It’s hard to detect those diseases manually and needs a lot of time and hard work. So, we introduce an agricultural and medical expert system based on machine vision that analyzes a picture acquired with a smartphone or portable device and classifies diseases to assist farmers in resolving the issue. We studied and used a convolutional neural network (CNN)-based transfer learning approach for identifying eggplant diseases in this paper. We have used various transfer learning models such as DenseNet201, Xception, and ResNet152V2. DenseNet201 had the highest accuracy of these models with 99.06%.

Keywords


Agro-medical expert system; DenseNet201; Eggplant diseases; Recognition; Transfer learning

Full Text:

PDF


DOI: https://doi.org/10.11591/eei.v11i5.3575

Refbacks

  • There are currently no refbacks.


Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

Bulletin of EEI Stats