ABSTRACT

Regrettably, fruit diseases may fail agricultural productivity and the global economy. Numerous studies have established that fruits are critical for sustaining optimal health. A healthy diet must include plenty of fruits if it is effective. The use of physical characteristics to diagnose fruit diseases is a novel and improved approach described in this study. The novel technique incorporates Artificial Neural Networks and Support Vector Machines. It maps images to their corresponding disease classifications using phenotypic traits such as texture, color, shape, fruit's hole structure, and physical makeup. Due to the usage of Artificial Neural Networks, this approach offers a few benefits over others in terms of detection and classification accuracy. They need far less preprocessing than conventional image classification approaches. There is a strong indication that the filters were analyzed by a network rather than a person. An essential benefit of this trait is that it is entirely independent of past knowledge and human effort. The Artificial Neural Network is combined with a Support Vector Machine to boost classification accuracy. The Proposed system can enhance disease identification and classification through more precise and automated approaches. Keywords: Fruit Diseases, Machine Leaning, Phenotyping, SVM, ANN
Publication date: 01/12/2021

https://ijbpas.com/pdf/2021/December/MS_IJBPAS_2021_DEC_SPCL_2026.pdf

Download PDF
https://doi.org/10.31032/IJBPAS/2021/10.12.2026