Analisis Perbandingan Metode Principal Component Analysis (PCA) Dan Linear Discriminant Analysis (LDA) Pada Daun Teh Menggunakan Artificial Neural Network (ANN)

RUTI, HERUWANTI (2023) Analisis Perbandingan Metode Principal Component Analysis (PCA) Dan Linear Discriminant Analysis (LDA) Pada Daun Teh Menggunakan Artificial Neural Network (ANN). Undergraduate Thesis thesis, Institut Teknologi Telkom Purwokerto.

	Text Cover.pdf Download (634kB)
	Text ABSTRAK.pdf Download (34kB)
	Text ABSTRACT.pdf Download (31kB)
	Text BAB I.pdf Download (106kB)
	Text BAB II.pdf Download (570kB)
	Text BAB III.pdf Download (706kB)
	Text BAB IV.pdf Restricted to Registered users only Download (328kB) | Request a copy
	Text BAB V.pdf Download (35kB)
	Text LAMPIRAN.pdf Restricted to Registered users only Download (229kB) | Request a copy

Abstract

Indonesia is also one of the eighth largest tea producing countries in the world, therefore the quality and quality of the tea produced is of utmost concern from harvest to post-harvest. When the climate is in the dry season, the growth of tea leaves will be slower and make the picking rotation system backward from the specified day. This makes the picking rotation system sometimes inappropriate. In this study, tea leaves were divided into 3 classes, namely leaves that were ready to be harvested (ripe), leaves that would be harvested in the near future (half ripe) and leaves that were not yet ready to be harvested (immature). The stages in this research are image acquisition, resizing and segmentation, and image transformation to HSI (Hue Saturation Intensity) and HSV (Hue Saturation Value) color features. Feature extraction by quantizing HSI and HSV colors. After obtaining the HSI and HSV feature values, the training image is classified using the Artificial Neural Network (ANN) method which will then be applied to the accuracy improvement method, namely Principal component analysis (PCA) and Linear Discriminant Analysis (LDA). Measurement and testing are carried out into 2 scenarios, in the first scenario using 90 training data, the other parameters use auto from mlpclassifer, training training data will be carried out 5 times. The second scenario is used to obtain test accuracy taken from test data with a total of 60 data, the other parameters use auto from mlpclassifer, the test accuracy is carried out 5 times. In this study, the accuracy of the PCA method was 72% and the LDA was 97%. Keywords: Tea, Hue Saturation Intensity (HSI), Hue Saturation Value (HSV), Artificial Neural Network (ANN), Principal component analysis (PCA) dan Linear Discriminant Analysis (LDA)

Actions (login required)

View Item