Wijaya, Ezra (2023) KLASIFIKASI JENIS IKAN CUPANG MENGGUNAKAN METODE GLCM DAN SVM. Undergraduate thesis, Universitas Pembangunan Nasional Veteran Jawa Timur.
|
Text (Cover)
19081010054-cover.pdf Download (911kB) | Preview |
|
|
Text (Bab 1)
19081010054-bab1.pdf Download (145kB) | Preview |
|
|
Text (Bab 2)
19081010054-bab2.pdf Download (1MB) | Preview |
|
![[img]](http://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text (Bab 3)
19081010054-bab3.pdf Restricted to Registered users only until 24 July 2026. Download (240kB) | Request a copy |
|
|
Text (Bab 4)
19081010054-bab4.pdf Restricted to Registered users only Download (2MB) | Request a copy |
|
|
Text (Bab 5)
19081010054-bab5.pdf Restricted to Registered users only Download (135kB) | Request a copy |
|
|
Text (Daftar Pustaka)
19081010054-daftarpustaka.pdf Download (89kB) | Preview |
Abstract
Betta fish often make it difficult for someone to determine types, in general the types of betta fish have similar body textures, fins and tail. Therefore the process of determining the type of betta fish is necessary done automatically by a computer system. So hopefully it can make it easier to determine the types of betta fish. In this study, the authors used the GLCM (Gray Level Co-Occurrence Matrix) method and the SVM (Support Vector Machine) method as the classifier. This study aims to create a software that can identify and classify betta fish through processing image and also to find out the accuracy of the GLCM and SVM methods that have not been ever done on betta fish objects. The total dataset is 1600 data image with data for each class totaling 400 image data, this dataset will used for training data. Meanwhile, data testing will take 20% training data or as many as 320. The results of this study indicate that the application can classify fish betta with fancy, placard, halfmoon, and crowntail types. By setting GLCM properties (contrast, correlation, dissimilarity, energy, and homogeneity) as well as distance (2) and angles (0.45). Produces SVM accuracy results with C=1 which differs for each kernel, the "linear" kernel accuracy is 72%, accuracy kernel "RBF" by 62%, kernel "sigmoid" accuracy by 41%, kernel accuracy "polynomial" of 55%
| Item Type: | Thesis (Undergraduate) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||||||
| Subjects: | T Technology > T Technology (General) | ||||||||||||
| Divisions: | Faculty of Computer Science > Departemen of Informatics | ||||||||||||
| Depositing User: | Ezra | ||||||||||||
| Date Deposited: | 24 Jul 2023 08:02 | ||||||||||||
| Last Modified: | 24 Jul 2023 08:02 | ||||||||||||
| URI: | http://repository.upnjatim.ac.id/id/eprint/15459 |
Actions (login required)
 |
View Item |