Segmentasi Semantik Berbasis Deeplabv3+ Untuk Pemantauan Pencemaran Sampah di Perairan Sungai
Abstract
Rivers have an important role for living things, including humans. The current condition of the river is worrying, river pollution results in poor water quality. This research develops two semantic segmentation models with DeepLabv3+ for optimal river condition monitoring. The dataset acquired with UAVs amounted to 95 images that have gone through data cleaning and manually annotated with binary values, then classified into garbage, and river categories. Both categories of models were trained separately using DeepLabv3+ Architecture. Based on the evaluation results, the best performance of the garbage model Dice Coeficient Loss obtained and IoU 68.89%. While the best performance of the river category model with Jaccard Loss achieved an IoU of 99.9%. Providing a garbage segmentation map that focuses on river waters, the binary segmentation results of each model are parallel integrated using an AND logic operation approach, garbage outside the river pixels is eliminated. Thus, displaying a segmentation map that focuses on garbage in river waters.
Keywords: Semantic segmentation; DeepLabv3+; River pollution; UAV imagery
Abstrak
Sungai mempunyai peran yang penting bagi makhluk hidup salah satunya manusia. Kondisi sungai saat ini mengkhawatirkan, pencemaran sungai mengakibatkan kualitas air menjadi buruk. Penelitian ini mengembangkan dua model segmentasi semantik dengan DeepLabv3+ untuk pemantauan kondisi sungai dengan optimal. Dataset diakuisisi dengan UAV berjumlah 95 citra telah malalui pembersihan data dan dianotasi manual dengan nilai biner, kemudian diklasifikasikan ke dalam kategori sampah, dan sungai. Kedua kategori model dilatih secara terpisah dengan menggunakan Arsitektur DeepLabv3+. Berdasarkan hasil evaluasi, performa terbaik dari model sampah Dice Coeficient Loss memperoleh dan IoU 68,89%. Sementara performa terbaik pada model kategori sungai dengan Jaccard Loss mencapai IoU 99,9%. Memberikan peta segmentasi sampah yang berfokus pada perairan sungai, hasil segmentasi biner dari setiap model secara paralel diintegrasikan menggunakan pendekatan operasi logika AND, sampah diluari piksel sungai dieliminasi. Sehingga menampilkan peta segmentasi yang berfokus pada sampah yang berada di perairan sungai.
Keywords
References
M. S. Harefa, S. Hidayat, K. Sisca, O. Luahambowo, R. Siregar, dan W. Asmara, “Analisis Pemanfaatan Air Sungai Bagi Kehidupan Masyarakat dari Aliran Air Terjun Sipiso Piso di Kecamatan Merek Kabupaten Karo,” Jurnal Ilmu Sosial, vol. 5, no. 5, pp. 26-37, 2024, doi: 10.6578/triwikrama.v5i5.6855.
H. Antara dkk., “English Title: Relationship between Humans and River Areas ‘Human Impacts on Cisadane River Water, West Java-Banten,’” Ethics and Law Journal: Business and Notary (ELJBN, vol. 2, no. 1, pp. 2988–1293, 2024, [Daring]. Tersedia pada: http://journals.ldpb.org/index.php/eljbn
Sukmawati, M. Anwar, dan Paharuddin, “Perilaku Masyarakat dalam Memanfaatkan Air Sungai Sebagai Air MCK Community Behavior in Using River Water as Toilet Water,” KEPO:Jurnal Keperawatan Profesional, vol. 3, no. 1, pp. 88–95, 2022, doi: 10.36590/v3i1.299.
F. Sugiester S, Y. W. Firmansyah, W. Widiyantoro, M. F. Fuadi, Y. Afrina, dan A. Hardiyanto, “Dampak Pencemaran Sungai di Indonesia Terhadap Gangguan Kesehatan : Literature Review,” Jurnal Riset Kesehatan Poltekkes Depkes Bandung, vol. 13, no. 1, pp. 120–133, Agu 2021, doi: 10.34011/juriskesbdg.v13i1.1829.
World Bank, “World Development Indicators: Population 2022,” 2023. Diakses: 11 Februari 2025. [Daring]. Tersedia pada: http://data.worldbank.org/data-catalog/world-development-indicators
SIPSN, “Capaian Kinerja Pengelolaan Sampah.” Diakses: 22 Maret 2025. [Daring]. Tersedia pada: https://sipsn.menlhk.go.id/sipsn/
A. Nggilu, N. Raffi Arrazaq, dan T. Thayban, “Dampak Pembuangan Sampah di Sungai Terhadap Lingkungan dan Masyarakat Desa Karya Baru,” Jurnal Normalita, vol. 10, no. 3, pp. 196–202, Sep 2022, Diakses: 28 Juni 2025. [Daring]. Tersedia pada: https://ejurnal.pps.ung.ac.id/index.php/JN/article/view/1795
P. Ony Andewi, K. Agus Seputra, K. Yota Ernanda Aryanto, L. Joni Erawati Dewi, dan F. Teknik dan Kejuruan, “Integrasi Teknologi Penginderaan Jauh Dan Machine Learning pada Web Gis untuk Pemetaan Potensi Banjir,” Jurnal Pendidikan Teknologi dan Kejuruan, vol. 22, no. 1, pp. 12–23, Jan 2025, doi: https://doi.org/10.23887/jptkundiksha.v22i1.87455.
A. Taryana, M. Rifa, E. Mahmudi, dan H. Bekti, “Analisis Kesiapsiagaan Bencana Banjir di Jakarta,” Jurnal Administrasi Negara), Februari, vol. 13, no. 2, pp. 302–311, 2022, Diakses: 28 Juni 2025. [Daring]. Tersedia pada: https://jurnal.unpad.ac.id/jane/article/ view/37997/16902
M. Gazali dan A. Widada, “Analisis Kualitas Dan Perumusan Strategi Pengendalian Pencemaran Air Sungai Bangkahulu Bengkulu,” Journal of Nursing and Public Health, vol. 9, no. 1, pp. 54–60, 2021, Diakses: 28 Juni 2025. [Daring]. Tersedia pada: https://jurnal.unived.ac.id/index.php/jnph/article/view/1441
T. Yang, S. Zhou, A. Xu, J. Ye, dan J. Yin, “An Approach for Plant Leaf Image Segmentation Based on YOLOV8 and the Improved DEEPLABV3+,” Plants, vol. 12, pp. 3438, Sep 2023, doi: 10.3390/plants12193438.
M. Tharani, A. W. Amin, F. Rasool, M. Maaz, M. Taj, dan A. Muhammad, “Trash Detection on Water Channels,” dalam Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), Springer Science and Business Media Deutschland GmbH, 2021, pp. 379–389. doi: 10.1007/978-3-030-92185-9_31.
M. Azizul Hakim, H. Emawati, dan D. Mujahiddin, “Pemanfaatan Pesawat Tanpa Awak Untuk Pemetaan dan Identifikasi Penutupan Lahan Pada Kawasan Hutan Pendidikan Unmul,” AGRIFOR, vol. 20, pp. 47-58, Mar 2021, doi: 10.31293/agrifor.v20i1.4900.
I. M. G. Sunarya, I Wayan Treman, dan Putu Zasya Eka Satya Nugraha, “Classification of Rice Growth Stage on UAV Image Based on Convolutional Neural Network Method,” Jurnal Nasional Pendidikan Teknik Informatika (JANAPATI), vol. 12, no. 1, pp. 146–155, Mei 2023, doi: 10.23887/janapati.v12i1.60959.
Y. and P. G. and S. F. and A. H. Chen Liang-Chieh and Zhu, “Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation,” dalam Computer Vision – ECCV 2018, M. and S. C. and W. Y. Ferrari Vittorio and Hebert, Ed., Cham: Springer International Publishing, 2018, pp. 833–851.
A. Ali, S. Acharjee, Md. M. Sk., S. Z. Alharthi, S. S. Chaudhuri, dan A. Akhunzada, “DWSD: Dense waste segmentation dataset,” Data Brief, vol. 59, p. 111340, 2025, doi: https://doi.org/10.1016/j.dib.2025.111340.
D. Bashkirova dkk., “ZeroWaste Dataset: Towards Deformable Object Segmentation in Cluttered Scenes,” dalam 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2022, pp. 21115–21125. doi: 10.1109/CVPR52688.2022.02047.
N. A. Muhadi, A. F. Abdullah, S. K. Bejo, M. R. Mahadi, dan A. Mijic, “Deep Learning Semantic Segmentation for Water Level Estimation Using Surveillance Camera,” Applied Sciences, vol. 11, no. 20, pp. 112-121, 2021, doi: 10.3390/app11209691.
M. W. A. Kesiman, I. M. D. Maysanjaya, I. M. A. Pradnyana, I. M. G. Sunarya, dan P. H. Suputra, “Profiling Balinese Dances with Silhouette Sequence Pattern Analysis,” dalam CENIM 2020 - Proceeding: International Conference on Computer Engineering, Network, and Intelligent Multimedia 2020, Institute of Electrical and Electronics Engineers Inc., Nov 2020, pp. 423–428. doi: 10.1109/CENIM51130.2020.9297893.
A. A. Paramartha, N. Marti, dan Y. E. Aryanto, “Comparison of classification model and annotation method for Undiksha’s official documents,” J Phys Conf Ser, vol. 1516, p. 012026, Apr 2020, doi: 10.1088/1742-6596/1516/1/012026.
N. L. P. R. Dewi, I. N. S. W. Wijaya, I. K. Purnamawan, dan N. W. Marti, “Model Classifer Judul Berita Pariwisata Indonesia Berdasarkan Sentimen,” Jurnal Teknologi Informasi dan Ilmu Komputer, vol. 11, no. 1, pp. 117–124, Feb 2024, doi: 10.25126/jtiik.20241117617.
P. H. Suputra, A. D. Sensusiati, M. D. Artaria, G. J. Verkerke, E. M. Yuniarno, dan I. K. E. Purnama, “Automatic 3D Cranial Landmark Positioning based on Surface Curvature Feature using Machine Learning,” Knowledge Engineering and Data Science, vol. 5, no. 1, pp. 27-38, Jun 2022, doi: 10.17977/um018v5i12022p27-40.
I. G. M. W. K. Widiantara, K. Y. E. Aryanto, dan I. M. G. Sunarya, “Application of the Learning Vector Quantization Algorithm for Classification of Students with the Potential to Drop Out,” Brilliance: Research of Artificial Intelligence, vol. 3, no. 2, pp. 262–269, Nov 2023, doi: 10.47709/brilliance.v3i2.3155.
S. Mohajerani dan P. Saeedi, “Cloud and Cloud Shadow Segmentation for Remote Sensing Imagery via Filtered Jaccard Loss Function and Parametric Augmentation,” IEEE J Sel Top Appl Earth Obs Remote Sens, vol. 14, pp. 4254–4266, Apr 2021, doi: 10.1109/JSTARS.2021.3070786.
C. H. Sudre, W. Li, T. K. M. Vercauteren, S. Ourselin, dan M. J. Cardoso, “Generalised Dice overlap as a deep learning loss function for highly unbalanced segmentations,” Deep learning in medical image analysis and multimodal learning for clinical decision support : Third International Workshop, DLMIA 2017, and 7th International Workshop, ML-CDS 2017, held in conjunction with MICCAI 2017 Quebec City, QC,..., vol. 2017, pp. 240–248, 2017, [Daring]. Tersedia pada: https://api.semanticscholar.org/CorpusID:21957663
P. Z. E. S. Nugraha, I. M. G. Sunarya, dan I. M. D. Maysanjaya, “Binary Semantic Segmentation of Dolphin on UAV Image Using U-Net,” dalam 2023 International Seminar on Intelligent Technology and Its Applications: Leveraging Intelligent Systems to Achieve Sustainable Development Goals, ISITIA 2023 - Proceeding, Institute of Electrical and Electronics Engineers Inc., 2023, pp. 728–733. doi: 10.1109/ISITIA59021.2023.10221152.
How To Cite This :
Refbacks
- There are currently no refbacks.
