CNN-Based Android App for Mulberry Leaf Disease Detection with NasNetMobile, Xception, and YOLO

Abstract

The *Bombyx mori* silkworms, which are important for making silk, in the main feed on mulberry leaves. Nevertheless, mulberry bushes are relatively prone to diseases which can spread quickly and purpose sizeable damage. Identifying diseases on huge farms with the aid of hand is a exhausting and time-consuming method. One option to this hassle is the usage of computer vision-based totally techniques for disease categorization and early diagnosis, which can cut production losses with the aid of ninety% or greater. Researchers on this take a look at categorized mulberry leaf samples as both healthy, rust-affected, or spot-affected after gathering statistics from areas in Bangladesh. The studies used classification algorithms, which took use of recent traits in device gaining knowledge of. The detection and classification overall performance of an ensemble version that protected “CNN, Xception, and NasNetMobile” was much higher. To locate illnesses, we used present day object detection algorithms like “YoloV5x6, YoloV8, and YoloV9” to look for anomalies at the leaves. The results exhibit the effectiveness of combining several models, imparting a strong answer to the trouble of ailment tracking in actual-time using an Android app based totally on convolutional neural networks (CNNs). Sustainable silk production may be accomplished with the help of this powerful and scalable approach, so as to resource farmers in reducing the effects of mulberry leaf illnesses.