River Cleanliness Measured on Lots of Waste Based on Digital Image Processing Using a Deep Convolutional Neural Network Penentuan Tingkat Kebersihan Sungai Berdasarkan Banyak Sampah Berbasis Pengolahan Citra Digital Menggunakan Deep Convolutional Neural Network

Main Article Content

Aditya Yuli Setyawan
Yosi Kristian


River environmental pollution by industrial waste and household waste is common today. In its function, the river is an ecosystem of various living things, including humans who often use rivers to support their daily activities. Along with the automatic development of the era, many technological inventions have been used to solve the problem of waste in rivers. This study discusses how to obtain a river cleanliness level using a Deep Convolutional Neural Network based on digital image processing as a method of detecting waste based on the amount of garbage in the river. To achieve this goal, several stages are required. The first stage begins with preprocessing, namely the process of making learning data which results in two categories, namely garbage and non-waste. The second stage is the process of recognizing objects on a computer that functions as a program dataset. The third stage is creating a selection area on the river surface which is given the term ROI (Region Of Intereset). The next step is calculating the average detected waste in pixels in the video from the average pixel results obtained as a reference to determine the level of river cleanliness. In this study, the level of river cleanliness was modeled uniformly into four classes, including the very dirty category with the percentage of waste detected 76% -100%, dirty with 51% -75% detected waste process, quite dirty with a percentage of detected waste 26% -50%, clean with a percentage of detected waste from 0% -25%. Furthermore, the calculation is simulated with a desktop-based application with the Python programming language.

Article Details



[1] Panella F at all. 2018. “Deep Learning and Image Processing for Automated Crack Detection and Defect Measurement in Underground Structures”. The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XLII-2.
[2] Socher, R., Huval, B., Bath, B., Manning, C. D., & Ng, A. Y. (2012). Convolutional-Recursive Deep Learning for 3D Object Classification. In Advances in Neural Information Processing Systems(pp. 656-664).
[3] Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2012). Imagenet Classification With Deep Convolutional Neural Networks. In Advances in neural information processing systems(pp. 1097-1105).
[4] Suartika, W., Wijaya, A.Y., dan Soelaiman, R. (2016). ”Klasifikasi Citra Menggunakan Convolutional Neural Network (CNN) pada Caltech 101”.Jurnal Teknik ITS Vol. 5, Hal. A65-A69.
[5] Mathworks. (2017). Convolutional Neural Network. Diambil kembali dari Mathworks: https://www.mathworks.com/discovery/convolutional-neural-network.html
[6] Rafael C. Gonzales, Richard E. Woods. 2002. “Digital Image Processing”. Tom Robbins Publisher. United States of America.
[7] Danukusumo,K.P. (2017). Implementasi Deep Learning Menggunakan Convolutional Neural Network Untuk Klasifikasi Citra Candi Berbasis GPU. Skripsi. Universitas Atma JayaYogyakarta.
[8] Canny, J. A. (1986) Computational Approach to Edge Detection. IEEE Transactions on Pattern Analysis and MachineIntelligence, 8, 679-698.
[9] Zhou, Ping, Wenjun Ye, Yaojie Xia, Qi Wang. (2011) An Improved Canny Algorithm for Edge Detection. Diambil dari:
[10] O’Shea K. & Nash R. (2015). “An Introduction to Convolutional Neural Networks”. arXiv:1511.08458v2 [cs.NE].
[11] Sulistyo, W., Bech, Y.R., Frans, F.Y., 2009, Analisis Penerapan Metode Median Filter Untuk Mengurangi Noise Pada Citra Digital, Konferensi Nasional Sistem dan Informatika 2009, Bali.