Amanda, Risna (2026) Implementasi Aplikasi Mobile Untuk Klasifikasi Jenis Daun Alpukat Menggunakan Mobilenetv3. Other thesis, Universitas Islam Riau.

Text skripsi_213510221_watermark.pdf - Published Version Restricted to Registered users only Download (2MB) | Request a copy

Abstract

The advancement of image processing and deep learning technologies offers significant potential for developing automated plant identification systems. This study aims to develop a classification system for avocado leaf types using the MobileNetV3 architecture and implement it in a mobile application built with Flutter. The primary dataset consists of 900 images of avocado leaves representing five classes: Aligator, Madu, Kendil, Super, and SW01. The dataset was divided into 80% training data, 10% validation data, and 10% testing data. Data augmentation was applied exclusively to the training set using five augmentation techniques, resulting in 720 images per class and increasing the total number of training images to 3,600. All images were normalized and resized to 224×224 pixels to match the input requirements of the MobileNetV3 model. The model was trained using a transfer learning approach with the Adam optimizer, a learning rate of 0.0001, and the Categorical Crossentropy loss function. Experimental results show that MobileNetV3-Large achieved the highest performance with an accuracy of 97%, while MobileNetV3-Small obtained an accuracy of 90%. Based on these results, MobileNetV3-Large was selected as the primary model for integration into the mobile application due to its superior classification performance and reliability. The implemented system enables realtime detection through the device camera or uploaded images. Overall, this study successfully produced an accurate, efficient, and mobile-ready avocado leaf classification model. The system is expected to assist users, farmers, and agricultural practitioners in identifying avocado varieties quickly and conveniently, as well as serve as a foundation for developing similar plant classification systems in the future.

Actions (login required)

View Item