Optimasi Segmentasi Bentuk untuk Mengklasifikasi Penyakit Daun Jagung Menggunakan Metode Faster R-CNN (Region Convolutional Neural Network) dan SVM (Support Vector Machine)

Pranajelita, Yuaini (2025) Optimasi Segmentasi Bentuk untuk Mengklasifikasi Penyakit Daun Jagung Menggunakan Metode Faster R-CNN (Region Convolutional Neural Network) dan SVM (Support Vector Machine). Undergraduate thesis, UPN Veteran Jawa Timur.

	Text (Cover) 21081010204.-Cover.pdf Download (2MB)
	Text (Bab 1) 21081010204.-bab1.pdf Download (164kB)
	Text (Bab 2) 21081010204.-bab2.pdf Restricted to Repository staff only until 20 July 2027. Download (604kB)
	Text (Bab 3) 21081010204.-bab3.pdf Restricted to Repository staff only until 20 July 2027. Download (633kB)
	Text (Bab 4) 21081010204.-bab4.pdf Restricted to Repository staff only until 20 July 2027. Download (2MB)
	Text (Bab 5) 21081010204.-bab5.pdf Download (219kB)
	Text (Daftar Pustaka) 21081010204.-daftarpustaka.pdf Download (189kB)

Abstract

Image classification is one of the main challenges in the field of digital image processing and computer vision. This study aims to classify diseases in corn leaves using the Faster R-CNN method, and compare it with a combination of the Faster R-CNN and SVM methods. The Faster R-CNN model is used to detect objects in images, while SVM is used to classify features that have been extracted from the image. The dataset used is sourced from Kaggle and data taken directly in corn fields. The research process involves the stages of data transformation, data augmentation, data split, and testing parameters on the SVM model such as kernel, C value, Gamma, and degree. Performance evaluation is carried out using accuracy, Precision, recall, and F1-Score. The test results show that the 60:20:20 data split scenario produces the best performance for both types of data. On direct data, the Faster R-CNN method gives the best results with an accuracy of 98.89%, Precision of 98.94%, recall of 98.89%, and F1-Score of 98.90. Meanwhile, in Kaggle data, the Faster R-CNN and SVM hybrid methods provide the highest results with accuracy, precision, recall, and F1-Score values of 98.88% each. The selection of parameters in SVM also greatly affects the classification results, where the poly kernel and parameter C = 0.1 produce the highest performance for direct data, while in Kaggle data the poly kernel with degree 2 and gamma 0.1 gives better results than other kernels. Keywords : Image classification, Faster R-CNN, SVM, feature extraction, object detection, hybird model, corn leaf disease

Item Type:

Thesis (Undergraduate)

Contributors: