LEAF CLASSIFICATION USING SVM AND CANNY EDGE DETECTION

TYASKO, YOSEPH STEFANO RAMADITYA (2021) LEAF CLASSIFICATION USING SVM AND CANNY EDGE DETECTION. Other thesis, Universitas Katolik Soegijapranata Semarang.

[img] Text (COVER)
19.K1.0021-YOSEPH STEFANO RAMADITYA TYASKO-COVER_a.pdf
Download (455kB)
[img] Text (BAB I)
19.K1.0021-YOSEPH STEFANO RAMADITYA TYASKO-BAB I_a.pdf
Download (114kB)
[img] Text (BAB II)
19.K1.0021-YOSEPH STEFANO RAMADITYA TYASKO-BAB II_a.pdf
Restricted to Registered users only
Download (116kB)
[img] Text (BAB III)
19.K1.0021-YOSEPH STEFANO RAMADITYA TYASKO-BAB III_a.pdf
Download (212kB)
[img] Text (BAB IV)
19.K1.0021-YOSEPH STEFANO RAMADITYA TYASKO-BAB IV_a.pdf
Download (125kB)
[img] Text (BAB V)
19.K1.0021-YOSEPH STEFANO RAMADITYA TYASKO-BAB V_a.pdf
Download (1MB)
[img] Text (BAB VI)
19.K1.0021-YOSEPH STEFANO RAMADITYA TYASKO-BAB VI_a.pdf
Download (109kB)
[img] Text (DAFTAR PUSTAKA)
19.K1.0021-YOSEPH STEFANO RAMADITYA TYASKO-DAPUS_a.pdf
Download (175kB)
[img] Text (LAMPIRAN)
19.K1.0021-YOSEPH STEFANO RAMADITYA TYASKO-LAMP_a.pdf
Download (183kB)

Abstract

Plants have evolved over the years and many changes happened to the structure and uses of the plants themselves, especially the leaves. Their leaves can be used for cooking, beverages, health, and herbal medicines. However, many leaf shapes are similar to each other, making it difficult to recognize the leaves and difficult to determine how to process them. For this reason, it is necessary to conduct research on leaf classification using edge patterns to identify plant species. From the above problems, it is recommended to create a program that classifies plants based on leaves. Canny edge detection is used for image processing and extracting leaf edge patterns. The classification process uses SVM using a trick kernel to process 3-dimensional data. By using the above method, the highest accuracy result is 89.37% with a cost value of 50. Combining SVM and canny detection can be used for classification, but the results of the combination are not as good as doing classification only with SVM.

Item Type: Thesis (Other)
Subjects: 600 Technology (Applied sciences)
Divisions: Faculty of Computer Science > Department of Informatics Engineering
Depositing User: Mrs Christiana Sundari
Date Deposited: 31 May 2023 06:38
Last Modified: 31 May 2023 06:38
URI: http://repository.unika.ac.id/id/eprint/31880

Actions (login required)

View Item View Item
Unika Repository is powered by EPrints 3 which is developed by the School of Electronics and Computer Science at the University of Southampton. More information and software credits.