Please use this identifier to cite or link to this item:
https://repositori.uma.ac.id/handle/123456789/24459
Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.advisor | Muhathir | - |
dc.contributor.author | Rifqi, Muhammad | - |
dc.date.accessioned | 2024-07-03T01:54:05Z | - |
dc.date.available | 2024-07-03T01:54:05Z | - |
dc.date.issued | 2024-03-28 | - |
dc.identifier.uri | https://repositori.uma.ac.id/handle/123456789/24459 | - |
dc.description | 55 Halaman | en_US |
dc.description.abstract | Tanaman kopi merupakan suatu jenis tanaman yang terdapat di daerah tropis dan subtropis yang membentang di sekitar garis equator, dan dapat hidup pada dataran rendah sampai dataran tinggi. Sebagai negara tropis, Indonesia merupakan salah satu negara penghasil kopi terbesar ke-3 di dunia setelah negara Vietnam dan Brazil. Ini adalah bagian dari bidang visi komputer dan pembelajaran mesin, di mana tujuannya adalah untuk mengajarkan komputer untuk mengenali dan membedakan objek atau pola dalam gambar serta proses untuk pengelompokan sejumlah pixel atau picture element pada sebuah citra menjadi kelas-kelas, pada masing-masing kelas mendiskripsikan suatu entitas yang mempunyai karakter agar dapat dikenali. Arsitektur ViT menonjol dalam metode klasifikasi citra melalui penggunaan teknik patching. Patching, yang melibatkan pembagian citra menjadi sejumlah kecil bagian seragam yang disebut sebagai "patches", menjadi fokus utama dalam pembelajaran komputer. Melalui proses ini, komputer mampu memahami dan menganalisis setiap patch untuk menghasilkan keputusan akhir Hasil akurasi dari model setelah diuji menggunakan data testing (100 citra per class) didapatkan hasil akurasi sebesar 97%, nilai presisi (precision) sebesar 98%, recall 97%, dan f1-score 97% pada masing-masing class. Coffee plants are a type of plant found in tropical and subtropical areas that stretch around the equator, and can live in the lowlands to highlands. As a tropical country, Indonesia is one of the 3rd largest coffee producing countries in the world after Vietnam and Brazil. It is part of the field of computer vision and machine learning, where the goal is to teach computers to recognize and distinguish objects or patterns in images as well as the process for grouping a number of pixels or picture elements in an image into classes, each class describing an entity that has characteristics to be recognized. ViT architecture stands out in image classification methods through the use of patching techniques. Patching, which involves dividing an image into a small number of uniform parts called "patches", is a major focus in computer learning. Through this process, the computer was able to understand and analyze each patch to produce a final decision. The accuracy results of the model after being tested using testing data (100 images per class) obtained an accuracy of 97%, a precision value of 98%, recall 97%, and fl-score 97% in each class. | en_US |
dc.language.iso | id | en_US |
dc.publisher | Universitas Medan Area | en_US |
dc.relation.ispartofseries | NPM;198160059 | - |
dc.subject | deep learning | en_US |
dc.subject | VIT | en_US |
dc.subject | klasifikasi | en_US |
dc.subject | deep learning | en_US |
dc.subject | classification | en_US |
dc.title | Klasifikasi Penyakit pada Daun Kopi Menggunakan Metode Vision Transformer (Vit) | en_US |
dc.title.alternative | Classification of Diseases on Coffee Leaves Using the Vision Transformer (Vit) Method | en_US |
dc.type | Skripsi Sarjana | en_US |
Appears in Collections: | SP - Informatic Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
198160059 - Muhammad Rifqi - Fulltext.pdf | Cover, Abstract, Chapter I, II, III, V, Bibliography | 1.07 MB | Adobe PDF | View/Open |
198160059 - Muhammad Rifqi - Chapter IV.pdf Restricted Access | Chapter IV | 476.77 kB | Adobe PDF | View/Open Request a copy |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.