Sistem Rejector Label Kemasan Susu Menggunakan Image processing Dengan Metode Support Vector Machine Pada Industri Pengolahan Susu

Yuliyanto, Jefrin (2021) Sistem Rejector Label Kemasan Susu Menggunakan Image processing Dengan Metode Support Vector Machine Pada Industri Pengolahan Susu. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

Text Buku Proyek Akhir_10311710000108_Jefrin Yuliyanto.pdf - Accepted Version Restricted to Repository staff only until 1 October 2023. Download (4MB) | Request a copy

Abstract

Pada pabrik pengolahan susu, terdapat beberapa permasalahan saat pemasangan label kemasan botol susu. Permasalahan tersebut adalah terjadinya gagal pemasangan atau sobeknya label pada saat dilakukan pemasangan label botol. Problem tersebut menjadi jobdesk tersendiri pada suatu pabrik, yang menyebabkan produk belum layak jual. Sehingga memerlukan perbaikan pada label dan produk harus di reject terlebih dahulu. Maka dari itu dirancang sebuah alat yang dapat mendeteksi kerusakan atau gagal pemasangan pada label kemasan susu menggunakan image processing dengan metode Gray Level Co-occurance Matrix (GLCM) sebagai pendeteksi texture dari objek dengan mengeluarkan nilai energy, entropy, homogeneity, dan contrast. Kemudian hasil pembacaan GLCM akan dilakukan klasifikasi menggunakan support vector machine (SVM) untuk memilah kondisi label botol. Jika salah label botol cacat maka botol akan di reject dan jika sempurna atau normal akan lanjut ke proses pengemasan. Hasil dari alat rejector label kemasan susu ini, Ketika mendeteksi label kemasan botol susu secara random menghasilkan nilai accuracy sebesar 85%, precission 90%, dan False Positive Rate 20%, Ketika dilakukan pengujian keseluruhan didapat hasil %error sebesar 20%.
=====================================================================================================
In milk processing plants, there are several problems when installing milk bottle packaging labels. The problem is the occurrence of failed installation or tearing of the label when the bottle label is installed. The problem becomes a separate job desk in a factory, which causes the product to be unfit for sale. So it requires improvement on the label and the product must be rejected first. Therefore, a tool is designed that can detect damage or failure of installation on milk packaging labels using image processing with the Gray Level Co-occurance Matrix (GLCM) method as a texture detector of objects by issuing energy, entropy, homogeneity, and contrast values. Then the results of the GLCM reading will be classified using a support vector machine (SVM) to sort out the condition of the bottle label. If the wrong bottle label is defective, the bottle will be rejected and if it is perfect or normal, it will continue to the packaging process. The results of this milk packaging label rejector tool, when detecting milk bottle packaging labels randomly produce an accuracy value of 85%, precision 90%, and a False Positive Rate of 20%. When the overall test is carried out, the result is an error of 20%.

Item Type:	Thesis (Undergraduate)
Uncontrolled Keywords:	image processing, Gray Level Co-occurance Matrix, support vector machine
Subjects:	T Technology > T Technology (General) > T57.8 Nonlinear programming. Support vector machine. Wavelets. Hidden Markov models. T Technology > TA Engineering (General). Civil engineering (General) > TA1637 Image processing--Digital techniques T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK2692 Inverters T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK2785 Electric motors, Induction. T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK3070 Automatic control
Divisions:	Faculty of Vocational > 36304-Automation Electronic Engineering
Depositing User:	Jefrin Yuliyanto
Date Deposited:	25 Aug 2021 23:32
Last Modified:	25 Aug 2021 23:32
URI:	http://repository.its.ac.id/id/eprint/89605

Actions (login required)

View Item