Klasifikasi Tanaman Beringin Berdasarkan Citra Daun Menggunakan Algoritma K-Nearest Neighbors

FAHIK, Yuliana Kiik (2024) Klasifikasi Tanaman Beringin Berdasarkan Citra Daun Menggunakan Algoritma K-Nearest Neighbors. Undergraduate thesis, Universitas Katolik Widya Mandira Kupang.

	Text ABSTRAK.pdf Download (389kB)
	Text BAB I.pdf Download (691kB)
	Text BAB II.pdf Download (850kB)
	Text BAB III.pdf Download (761kB)
	Text BAB IV.pdf Download (319kB)
	Text BAB V.pdf Download (1MB)
	Text BAB VI.pdf Download (503kB)
	Text DAFTAR PUSTAKA.pdf Download (149kB)

Abstract

Image classification is an important technique in digital image processing which aims to group objects or data into appropriate categories based on their features. There are several types of banyan plants, namely, white banyan, dollar banyan, kimeng banyan and elegant banyan. This research aims to classify types of banyan using the K-Nearest Neighbors (KNN) algorithm. The KNN approach is a non-parametric method that utilizes information from K-Nearest Neighbors to classify new objects. In this research, the stage of collecting data on images of banyan plants from reliable sources was carried out. These images were then digitally processed to extract important features such as texture, shape and color from each banyan. Next, the image dataset is used as training data to train the KNN model. In the testing stage, the previously unseen banyan image will be classified using a KNN model that has been trained with evaluation metrics such as accuracy, precision, recall and F1-score. The experimental results show that the KNN-based banyan plant image classification method is able to classify banyan images with a good level of success, and provides a basis for further development of image classification techniques in the context of agriculture and plant research. In this research, KNN classified 4 types of banyan images using the orange data mining tools as a tool to carry out the data mining process. The results of the classification carried out obtained quite high accuracy, which was at number of folds 20 with the classification of dollar banyan leaves getting 100% accuracy, white banyan leaves getting 95% accuracy, elegant banyan leaves getting 98.75% accuracy and kimeng banyan leaves getting accuracy 100%.

Actions (login required)

View Item