Klasifikasi Sampah Daur Ulang Menggunakan Support Vector Machine Dengan Fitur Local Binary Pattern

Abstract

Garbage is one of the problems that always arise in Indonesia and even in the world. Increasingly, the production of waste is increased along with the increase in population and consumption. Therefore, need a prevention to stop wasting or producing garbage through recycle. This research do garbage recycle classification of cardboard, glass, metal, paper and plastic by using Local Binary Pattern (LBP) texture feature extraction methode and Support Vector Machine (SVM) as classification methode. For examination technic and dataset distribution is using K-Fold Cross Validation methode type Leave One Out (LOO). From examination result had been done were using fold 5 until fold 10. Polynomial kernel get highest accuracy result from every fold used with mean point 87.82%. Based on SVM classification examination result whether linear kernel, polynomial nor gaussian by using fold 5 until fold 10. The best accuracy point for cardboard garbage is 96.01%. For glass garbage, the best accuracy point is 90.62%. Then, metal garbage get the best accuracy point 89.72%. While paper garbage with highest accuracy point 96.01%. And plastic garbage with highest accuracy point 87.64%.

Image Processing Using Morphology on Support Vector Machine Classification Model for Waste Image

“…Figure 10 is the result of tests carried out using machine learning that has been trained with training data and tested with test data. The results of this study compared to previous studies by Leonardo and Chu have several differences [10,11]. The first difference is that the data collection used is a dataset collected on Kaggle [21].…”

“…Supervised learning is a type of machine learning that will be created because the category of types of waste has been determined in the rules formed in machine learning so that the machine no longer needs to cluster data into categories that the machine determines [9]. Related work to this research is research conducted by Leonardo [10]. The research explains that SVM is one of the image classification models that can be used properly.…”

The Application of the Fletcher-Reeves Algorithm to Predict Spinach Vegetable Production in Sumatra

“…The production data of spinach plants on the island of Sumatra (Table 1) were normalized using the above formula and then divided into two groups, namely the training dataset and the test dataset [11,12]. The training dataset is from 2015-2016 (X1-X2), and the target year is 2017 (Y1) (Table 2).…”

Identifikasi Jenis Sampah Secara Otomatis Menggunakan Metode Convolutional Neural Network (Cnn)

“…Penelitian sebelumnya yang dilakukan oleh Leonardo et al, [8] untuk melakukan klasifikasi sampah daur ulang menggunakan Support Vector Machine dengan fitur Local Binnary Pattern. Penelitian ini menggunakan dataset berisi citra sampah yang dapat didaur ulang dibagi menjadi enam kelas masing-masng sekitar 400-500 citra dengan ukuran 400x384 pixel menghasilkan tingkat akurasi kernel linear terbaik yaitu 87,63%.…”