KLASIFIKASI JENIS RETAKAN PADA BETON MENGGUNAKAN ALGORITMA HISTOGRAM OF GRADIENTS DAN MACHINE LEARNING

MAKRUFIAH SAKATRI (2022) KLASIFIKASI JENIS RETAKAN PADA BETON MENGGUNAKAN ALGORITMA HISTOGRAM OF GRADIENTS DAN MACHINE LEARNING. S1 thesis, Universitas Muhammadiyah Yogyakarta.

[thumbnail of Halaman Judul] Text (Halaman Judul)
Halaman Judul.pdf

Download (2MB)
[thumbnail of Lembar Pengesahan] Text (Lembar Pengesahan)
Lembar Pengesahan.pdf
Restricted to Registered users only

Download (3MB)
[thumbnail of Abstrak] Text (Abstrak)
Abstrak.pdf
Restricted to Registered users only

Download (93kB)
[thumbnail of Bab I] Text (Bab I)
Bab I.pdf

Download (491kB)
[thumbnail of Bab II] Text (Bab II)
Bab II.pdf
Restricted to Registered users only

Download (687kB)
[thumbnail of Bab III] Text (Bab III)
Bab III.pdf
Restricted to Registered users only

Download (3MB)
[thumbnail of Bab IV] Text (Bab IV)
Bab IV.pdf
Restricted to Registered users only

Download (9MB)
[thumbnail of Daftar Pustaka] Text (Daftar Pustaka)
Daftar Pustaka.pdf
Restricted to Registered users only

Download (112kB)
[thumbnail of Naskah Publikasi] Text (Naskah Publikasi)
Naskah Publikasi.pdf
Restricted to Registered users only

Download (424kB)
[thumbnail of Full Text] Text (Full Text)
Full Text.pdf
Restricted to Repository staff only

Download (18MB)

Abstract

Concrete crack frequently is a serious problem occurs around building structure and road structure. There are requirements to be an early inspection before the concrete used. This research builds a model to classify type concrete surface by Image Processing. In this era, technology always continues to develop a Machine Learning is Artificial Intelligence to help human’s work from any sector such as medical, economic, agriculture and other. Concrete structure inspection can be worked by builds Machine Learning system within concrete crack surface image as input data. The model system in this research using HOG feature extraction and SVM-KNN classification. The HOG extraction feature process using 50 60 cell size then it worked as shape feature and object detection as numerical extraction. Accuracy training and performance matrix on testing will be set as parameter model. The result shows model on HOG 1 using Medium Guassian SVM the training accuracy 89%, HOG 2 using Quadratic SVM the training accuracy 82.40% and HOG 3 using Quadratic SVM the training accuracy 90.20%.

Item Type: Thesis (S1)
Uncontrolled Keywords: CONCRETE STRUCTURE, IMAGE PROCESSING, SVM, KNN, EXTRACTION FEATURE HOG.
Divisions: Fakultas Teknik > Teknik Elektro S1
Depositing User: M. Erdiansyah
Date Deposited: 05 Sep 2022 03:08
Last Modified: 05 Sep 2022 03:08
URI: https://etd.umy.ac.id/id/eprint/34101

Actions (login required)

View Item
View Item