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%.
Highlights
Sampah merupakan salah satu masalah yang selalu muncul di Negara Indonesia bahkan didunia
Kemudian penelitian yang dilakukan [4] sistem deteksi mata uang kertas berdasarkan fitur gabungan speeded up robust feature (SURF) dan local binary pattern (LBP) dengan menggunakan klasifikasi support vector machine (SVM)
Beberapa studi yang mengadopsi validasi leave-one out cross untuk mengevaluasi performa dari penggolongan metode klasifikasi ketika jumlah instans kedalam data training dan testing tidak ada, poin estimasi keakuratan untuk data yang diberikan adalah konstan
Summary
Sampah merupakan salah satu masalah yang selalu muncul di Negara Indonesia bahkan didunia. Menurut [17] yang melakukan penelitian tentang klasifikasi sampah daur ulang menggunakan metode support vector machine (SVM) dengan fitur scale-invariant feature transform (SIFT) dan convolutional neural network (CNN). Hasil penelitian menunjukkan bahwa SVM berkinerja lebih baik dari CNN yang tentunya telah membuka jalan untuk meningkatkan akurasi dan ketepatan klasifikasi sampah daur ulang. Kemudian penelitian yang dilakukan [4] sistem deteksi mata uang kertas berdasarkan fitur gabungan speeded up robust feature (SURF) dan local binary pattern (LBP) dengan menggunakan klasifikasi support vector machine (SVM). Pada tahap awal citra di preprocessing terlebih dahulu guna menghasilkan kualitas yang baik setelah itu dilakukan ekstraksi fitur gabungan menggunakan SURF dan LBP kemudian baru diklasifikasi SVM.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.