Aplikasi Computer Vision Untuk Mengenali Kandungan Bahan Babi Pada Makanan Yang Tertera Dalam Komposisi Makanan Pada Kemasan

Kholis, Nur (2020) Aplikasi Computer Vision Untuk Mengenali Kandungan Bahan Babi Pada Makanan Yang Tertera Dalam Komposisi Makanan Pada Kemasan. Undergraduate thesis, Politeknik Negeri Jember.

	Text (Abstract) 08. ABSTRACT.pdf - Submitted Version Available under License Creative Commons Attribution Share Alike. Download (88kB)
	Text (Bab 1 Pendahuluan) 15. BAB 1. PENDAHULUAN.pdf - Submitted Version Available under License Creative Commons Attribution Share Alike. Download (107kB)
	Text (Daftar Pustaka) 20. DAFTAR PUSTAKA.pdf - Submitted Version Available under License Creative Commons Attribution Share Alike. Download (96kB)
	Text (Laporan Lengkap) E41190091 _LAPORAN LENGKAP.pdf Restricted to Registered users only Download (21MB)

Abstract

Packaged foods provide information about the composition of the food. Some of the food in circulation does not only come from Indonesia but there are several foods that come from various countries, so the officers of the Food and Drug Supervisory Agency (BPOM) in carrying out their duties must supervise and examine the composition of the food one by one. In the composition there are several food ingredients written with a certain code so that it is difficult to determine whether the food ingredients come from pigs or not. To solve this problem, an Android-based Computer Vision application was created using the Firebase Machine Learning Kit (MLKit) text recognition with the addition of the Cosine Similarity and Levenshtein methods. From the test results, the Levenshtein method achieves an average accuracy of 70%, a precision of 68% and a recall of 75%. Meanwhile, the Cosine Similarity method has an average accuracy value of 62.5%, precision of 63.5%, recall of 55%. It is hoped that the application can help make it easier for BPOM officers to check the composition of food, especially foods containing pork.

Actions (login required)

View Item