PENERAPAN MODEL CONVNEXT DALAM MENGKLASIFIKASIKAN PENYAKIT DAUN TANAMAN KENTANG DI LINGKUNGAN YANG TIDAK TERKENDALI (POTATO LEAF DISEASE IN UNCONTROLLED ENVIRONMENT)

Tirta Agung Jati, NIM.: 20106050001 (2024) PENERAPAN MODEL CONVNEXT DALAM MENGKLASIFIKASIKAN PENYAKIT DAUN TANAMAN KENTANG DI LINGKUNGAN YANG TIDAK TERKENDALI (POTATO LEAF DISEASE IN UNCONTROLLED ENVIRONMENT). Skripsi thesis, UIN SUNAN KALIJAGA YOGYAKARTA.

Preview	Text (PENERAPAN MODEL CONVNEXT DALAM MENGKLASIFIKASIKAN PENYAKIT DAUN TANAMAN KENTANG DI LINGKUNGAN YANG TIDAK TERKENDALI (POTATO LEAF DISEASE IN UNCONTROLLED ENVIRONMENT)) 20106050001_BAB-I_IV-atau-V_DAFTAR-PUSTAKA.pdf - Published Version Download (2MB) | Preview
	Text (PENERAPAN MODEL CONVNEXT DALAM MENGKLASIFIKASIKAN PENYAKIT DAUN TANAMAN KENTANG DI LINGKUNGAN YANG TIDAK TERKENDALI (POTATO LEAF DISEASE IN UNCONTROLLED ENVIRONMENT)) 20106050001_BAB-II_sampai_SEBELUM-BAB-TERAKHIR.pdf - Published Version Restricted to Registered users only Download (3MB) | Request a copy

Abstract

Potato (Solanum tuberosum L.) significantly impacting the global economy especially in Dieng, Indonesia. Rapid and accurate identification of potato plant diseases is crucial for farmers to avoid economic losses, given their limited knowledge about these diseases. Artificial Intelligence (AI) has demonstrated remarkable capabilities in processing and analyzing images for various applications, including image classification. Currently, research results for identifying potato plant diseases, primarily utilizing potato leaf images from the Potato Leaf Disease Dataset in Uncontrolled Environment, still show unsatisfactory results. The accuracy achieved by models from previous studies is still less than 75%. This study utilizes deep learning techniques, especially convolutional neural networks (CNNs) by proposing the use of the ConvNeXt model with transfer learning and fine-tuning techniques. The model is evaluated through multiclass statistical analysis based on accuracy, precision, recall, and F1-Score. The results show that the proposed ConvNeXt-Large model outperforms existing models with an accuracy of 84.57%, a precision of 0.8295, a recall of 0.7733, and an F1-Score of 0.8004.

Share this knowledge with your friends :

Actions (login required)

View Item