Klasifikasi Daging Sapi Berbasis Electronic Nose

Daiva, Aldhiaz Fathra (2018) Klasifikasi Daging Sapi Berbasis Electronic Nose. Undergraduate thesis, Institut Teknologi Sepuluh Nopember.

[img]
Preview
Text
5113100006-Undergraduate_Thesis.pdf - Accepted Version

Download (5MB) | Preview

Abstract

Daging merupakan salah satu konsumsi utama manusia. Namun dalam praktik jual beli masih terdapat pencampuran jenis daging untuk mendapatkan keuntungan yang lebih secara curang. Terlebih pencampuran menggunakan daging Babi yang dalam ajaran Islam merupakan makanan yang dilarang. Dengan dilakukannya penelitian ini, kami merancang dan mengusulkan sistem sederhana berbiaya rendah yang dapat mendeteksi kemurnian daging berdasarkan bau yang dideteksi dengan mengukur konsentrasi gas yang dikeluarkan daging melalui electronic nose yang dirakit sendiri. Sistem ini dibangun melalui enam tahap: pembuatan perangkat keras electronic nose menggunakan sensor gas dan Arduino; pengambilan sampel data; praproses data menggunakan discrete wavelet transform untuk mengurangi noise; ekstraksi fitur statistik; seleksi fitur menggunakan chi-squared statistic weighting; dan klasifikasi menggunakan metode kNN, decision tree, naïve bayes, dan SVM. Hasil penelitian menunjukkan bahwa sistem ini dapat membedakan daging sapi yang dicampur dengan daging babi dengan perbandingan 0%, 10%, 25%, 50%, 75%, 90%, dan 100% dengan akurasi 97,86% menggunakan algoritma kNN. =============================================================================== Meat is one of the mainly consumed food for human. However in the real practice, meat adultery by mixing other type of meat is still occurring. Especially mixing using pork meat which in Islamic view is a prohibited food. This practice is done to fraudulently gain more income. In this research, we develop and propose an easy to use, low cost system to classify wether the meat is a pure beef or not. The system is developed through six main stages: developing of electronic nose hardware using gas sensors and Arduino, gathering ground truth data for the training set and evalution, data preprocessing by utilizing discrete wavelet transform for denoising, statistical features extraction, feature selection using chi-squared weighting, and classification using kNN, decision tree, naïve bayes, and SVM. The experiment results show that this system can classify pork adulteration in a beef with percentage of 0% 10%, 25%, 50%, 75%, 90%, and 100% with performance of 97.86% of accuracy using kNN algorithm.

Item Type: Thesis (Undergraduate)
Additional Information: RSIf 681.2 Dai k
Uncontrolled Keywords: klasifikasi, daging, electronic nose, Arduino, pemrosesan sinyal
Subjects: T Technology > T Technology (General) > T57.5 Data Processing
T Technology > T Technology (General) > T58.6 Management information systems
Divisions: Faculty of Information and Communication Technology > Informatics > 55201-(S1) Undergraduate Thesis
Depositing User: Aldhiaz Fathra Daiva
Date Deposited: 05 Nov 2020 06:24
Last Modified: 05 Nov 2020 06:24
URI: https://repository.its.ac.id/id/eprint/59252

Actions (login required)

View Item View Item