Skripsi
PENERAPAN SISTEM SORTIR SAMPAH KERTAS OTOMATIS MENGGUNAKAN METODE NAÏVE BAYES SECARA REAL-TIME
Paper waste is a type of waste that continues to increase in number so that it requires proper management so that it can be recycled optimally. One of the efforts to support the recycling process is to sort paper waste by type. This research aims to develop a real-time paper waste classification system based on digital image processing using HSV color features and contour features that cover areas and perimeters. The classification method used is Naïve Bayes to determine the level of accuracy of the system in distinguishing paper types. The research was carried out through several stages, namely dataset collection, system design, digital image processing, smart cropping, feature extraction, and the classification and evaluation process of method performance. The dataset used amounted to 300 images consisting of three types of paper, namely Uncoated Woodfree, Kraft Paper, and Bleached Paperboard, with a share of 70% training data and 30% testing data. The system is designed in the form of a prototype consisting of a belt conveyor, web camera, mini computer, mini monitor, and wireless keyboard. The test results showed that the Naïve Bayes method was able to produce a classification accuracy of 77%, with the highest posterior probability value in the Kraft Paper class of 0.0716. The results of the study show that an image-based automatic paper waste sorting system can be implemented to classify paper types
| Title | Edition | Language |
|---|---|---|
| ANALISIS KINERJA METODE K-NEAREST NEIGHBORS UNTUK PENGEMBANGAN SISTEM OTOMASI KLASIFIKASI SAMPAH KERTAS SECARA REAL-TIME | id |