Klasifikasi Hama Dan Penyakit Tanaman Kapas Menggunakan Metode Gray Level Co-Occurrence Matrix Dan Multilayer Perceptron

AWANDI, NADHIF MAHARDIKA (2023) Klasifikasi Hama Dan Penyakit Tanaman Kapas Menggunakan Metode Gray Level Co-Occurrence Matrix Dan Multilayer Perceptron. Undergraduate thesis, UNIVERSITAS PEMBANGUNAN NASIONAL “VETERAN” JAWA TIMUR.

[img]
Preview
Text (Cover)
19081010064-cover.pdf

Download (663kB) | Preview
[img]
Preview
Text (Bab 1)
19081010064-bab1.pdf

Download (84kB) | Preview
[img] Text (Bab 2)
19081010064-bab2.pdf
Restricted to Registered users only until 31 May 2025.

Download (469kB)
[img] Text (Bab 3)
19081010064-bab3.pdf
Restricted to Registered users only until 31 May 2025.

Download (597kB)
[img] Text (Bab 4)
19081010064-bab4.pdf
Restricted to Registered users only until 31 May 2025.

Download (801kB)
[img]
Preview
Text (Bab 5)
19081010064-bab5.pdf

Download (73kB) | Preview
[img]
Preview
Text (Daftar Pustaka)
19081010064-daftarpustaka.pdf

Download (75kB) | Preview

Abstract

The Textile and Textile Product (TPT) industry in Indonesia has experienced a fairly good increase in the last few years. However, problem the supply of cotton fiber raw materials is still an obstacle that needs to be overcome by the industry. The solution that can be done to increase productivity is to prevent and control pests and diseases that attack cotton plants. To overcome the problem of identifying pests and diseases in cotton plants, image processing methods can be used, such as feature extraction of an object so identification can be carried out. This research utilizes the features of the Gray Level Co-occurrence Matrix (GLCM) method and the Multilayer Perceptron (MLP) method for the classification process. Based on the test results on the test data, the highest accuracy value is obtained by a combination of the Color Moment feature and the GLCM feature with an angle of 0⁰ and the MLP architecture using the tanh activation function and 1024 perceptrons in the hidden layer with an accuracy of 90%. Whereas in testing with images obtained from Google Image, the accuracy value obtained is 73%.

Item Type: Thesis (Undergraduate)
Contributors:
ContributionContributorsNIDN/NIDKEmail
Thesis advisorNUGROHO, BUDINIDN0707098003budinugroho.if@upnjatim.ac.id
Thesis advisorAKBAR, FAWWAZ ALINIDN0017039201fawwaz_ali.fik@upnjatim.ac.id
Subjects: T Technology > T Technology (General)
Divisions: Faculty of Computer Science > Departemen of Informatics
Depositing User: Mr. Nadhif Mahardika Awandi
Date Deposited: 06 Jun 2023 05:00
Last Modified: 06 Jun 2023 05:00
URI: http://repository.upnjatim.ac.id/id/eprint/14299

Actions (login required)

View Item View Item