Hikmatiyar, Dzaky Adla (2024) Deteksi Barisan Padi pada Sawah Jajar "Legowo" Berbasis YOLO dan Transformasi Hough. Other thesis, Institut Teknologi Sepuluh Nopember.
Text
5009201005-Undergraduate_Thesis.pdf - Accepted Version Restricted to Repository staff only until 1 October 2026. Download (4MB) | Request a copy |
Abstract
Padi merupakan salah satu komoditas yang sangat penting di Indonesia dimana hasil dari padi yaitu beras merupakan makanan pokok dari masyarakat Indonesia. Di tengah dinamika pertanian padi, sistem tanam jajar legowo muncul sebagai inovasi yang menarik perhatian. Namun, pada saat ini, aktivitas seperti penanggulangan gulma dan penanaman padi di Indonesia masih dilakukan menggunakan tenaga manusia. Hal ini kurang efisien dimana kegiatan ini seharusnya dapat dilakukan secara otomatis menggunakan smart tractor. Dalam merancang sebuah kendaraan yang bersifat autonomous, salah satu hal yang perlu dipertimbangkan adalah bagaimana kendaraan tersebut dapat bergerak tetap di dalam jalurnya. Untuk memenuhi kondisi tersebut, dapat dibuat sebuah pendeteksi barisan padi. Deteksi barisan padi akan dilakukan menggunakan YOLOv4 dan transformasi Hough. YOLOv4 digunakan untuk mendeteksi koordinat posisi padi. Hasil deteksi padi kemudian diolah menggunakan transformasi Hough sehingga dapat menghasilkan garis virtual yang menggambarkan barisan padi. Model deteksi barisan padi telah berhasil dibuat dan memperoleh performa yaitu kecepatan deteksi sebesar 21 fps dengan akurasi deteksi sudut kemiringan barisan padi sebesar 97.5%.
==============================================================================================================================
Paddy is one of the most important commodities in Indonesia where the result of it which is rice is the one of the primary foods of Indonesian residence. Amid the dynamics of rice farming, the "Jajar Legowo" planting system emerged as an innovation that attracted attention. However, currently, activities such as controlling weeds and planting the rice plant Indonesia are still carried out using human power. This is less efficient because this activity should be carried out automatically using a smart tractor. When designing an autonomous vehicle, one of the things that needs to be considered is how the vehicle can move steadily along its path. To meet these conditions, a paddy rows detector can be designed. Paddy rows detection will be carried out using YOLOv4 and Hough transformation. YOLOv4 is used to detect paddy position coordinates. The paddy detection results are then processed using the Hough transformation to produce virtual lines depicting rows of paddy. The paddy rows detection model has been successfully designed and achieved detection speed of 21 fps with paddy rows angle detection accuracy of 97.5%.
Item Type: | Thesis (Other) |
---|---|
Uncontrolled Keywords: | Hough transform, legowo paddy field, paddy rows detection, YOLO, deteksi barisan padi, sawah jajar legowo, transformasi Hough |
Subjects: | Q Science > QA Mathematics > QA336 Artificial Intelligence S Agriculture > SB Plant culture > SB191.R5 Rice farming T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK7882.P3 Pattern recognition systems |
Divisions: | Faculty of Industrial Technology and Systems Engineering (INDSYS) > Physics Engineering > 30201-(S1) Undergraduate Thesis |
Depositing User: | Dzaky Adla Hikmatiyar |
Date Deposited: | 29 Aug 2024 03:04 |
Last Modified: | 29 Aug 2024 03:04 |
URI: | http://repository.its.ac.id/id/eprint/109257 |
Actions (login required)
View Item |