Deteksi Limbah Organik Berbasis Deep Learning Untuk Proses Produksi Pupuk Bio-Organik Petrokimia
DOI:
https://doi.org/10.35870/ljit.v4i2.8545Keywords:
Organic Waste Detection, Deep Learning, Enhanced YOLOv5, Partial Layer Freezing, Bio-Organic Fertilizer, Petrochemical Industry.Abstract
The increasing volume of organic waste in Indonesia, combined with the reliance on manual raw material selection for bio-organic fertilizer production, has become a major challenge in the modern petrochemical industry due to its susceptibility to human error. This study aims to design, evaluate, and optimize a deep learning-based organic waste detection model using the YOLOv5 algorithm to improve the accuracy and efficiency of automated raw material sorting. The research employed a quantitative experimental approach using the Research and Development (R&D) method. Model performance was enhanced through the Enhanced YOLOv5 approach by implementing Partial Layer Freezing on the backbone (freeze = [0]), hyperparameter optimization, and data augmentation. Performance evaluation was conducted by comparing the Baseline YOLOv5s model with several Enhanced YOLOv5 variants trained at different epochs (10, 15, 25, and 35 epochs) using Precision, Recall, mAP50, mAP50-95, and inference time as evaluation metrics. The experimental results indicate that the Enhanced YOLOv5 Epoch 25 model achieved the optimal performance (sweet spot), with a Precision of 99.53%, Recall of 100%, mAP50 of 99.50%, and an inference speed of 8.59 ms per image (approximately 116 FPS) while maintaining an efficient model size of 17.66 MB. This approach improved inference speed by 38.68% compared to the Baseline YOLOv5s model (14.01 ms) without compromising detection accuracy. Therefore, the Enhanced YOLOv5 Epoch 25 model is highly recommended for integration into conveyor-based automated raw material sorting systems in the petrochemical industry.
Downloads
References
Ahmad, B., Khan, M., & Ali, T. (2024). Mitigating Background Noise and False Positives in Real-Time Object Detection Systems. IEEE Access, 12, 32145–32156. https://doi.org/10.1109/ACCESS.2024.3365892
Akmal, B. K., Lestari, W., & Pradana, A. I. (2026). Real-Time Waste Detection System Using YOLOv12 with Transfer Learning. Journal of Applied Informatics and Computing (JAIC), 10(1). https://jurnal.polibatam.ac.id/index.php/JAIC/article/view/12050
Alvarez, M., Rodriguez, D., & Perez, J. (2022). Identifying the Optimal Training Sweet Spot for Partial Layer Freezing and Augmentation in Industrial Vision Systems. IEEE Transactions on Industrial Informatics, 18(12), 8910–8920. https://doi.org/10.1109/TII.2022.3175401
Alvarez, M., Rodriguez, D., & Perez, J. (2023). Synergistic Effects of Partial Layer Freezing and Augmentation for Rapid Convergence in Industrial Inspection. Journal of Real-Time Image Processing, 20(4), 78. https://doi.org/10.1007/s11554-023-01332-9
Badan Pusat Statistik. (2024). Statistik Lingkungan Hidup Indonesia 2024.
Bochkovskiy, A., Wang, C.-Y., & Liao, H.-Y. M. (2020). YOLOv4: Optimal Speed and Accuracy of Object Detection. ArXiv Preprint ArXiv:2004.10934. https://arxiv.org/abs/2004.10934
Chen, L., Zhang, W., & Zhao, J. (2022). Impact Analysis of Partial Layer Freezing and Data Augmentation on Early-Stage Convergence in Deep Object Detection. IEEE Transactions on Industrial Informatics, 18(10), 6812–6822. https://doi.org/10.1109/TII.2022.3152401
Chen, L., Zhang, W., & Zhao, J. (2023). Accelerating Real-Time Inference via Partial Layer Freezing in Edge-Based Vision Inspection Systems. Expert Systems with Applications, 230, 120550. https://doi.org/10.1016/j.eswa.2023.120550
Chen, X., Xu, J., & Zhou, P. (2023). Confidence Threshold Adaptation and Bounding Box Refinement in Multi-Epoch Deep Object Detection. Neurocomputing, 535, 215–227. https://doi.org/10.1016/j.neucom.2023.03.012
Dobrzycki, A. D., Bernardos, A. M., & Casar, J. R. (2025). Analysis of Layer-Freezing Strategies for Enhanced Transfer Learning in YOLO Architectures. Mathematics, 13(15), 2539. https://doi.org/10.3390/math13152539
Freepik. (2024). Neural Network and Deep Learning Visualization. https://www.freepik.com/free-photo/representation-user-experience-interface-design_36034114.htm
Garcia, J., Martinez, M., & Lopez, D. (2022). Confusion Matrix Analysis of Standard YOLOv5 Architecture for Automatic Waste Sorting. Journal of Cleaner Production, 345, 131102. https://doi.org/10.1016/j.jclepro.2022.131102
Guo, Y., Chen, Z., Li, W., & Zhang, D. (2024). Mitigating Background False Positive Spikes in Intermediate Epoch Training for Industrial Vision Systems. Expert Systems with Applications, 241, 122580. https://doi.org/10.1016/j.eswa.2023.122580
Hassan, M., Ibrahim, A., & Youssef, K. (2024a). Deploying Lightweight Object Detection Models for Automated Bio-Organic Fertilizer Feedstock Sorting. Waste Management, 181, 98–109. https://doi.org/10.1016/j.wasman.2024.03.022
Hassan, M., Ibrahim, A., & Youssef, K. (2024b). Precision Visual Sorting of Organic Waste Materials for Bio-Fertilizer Feedstock Processing. Waste Management, 179, 112–123. https://doi.org/10.1016/j.wasman.2024.03.008
He, K., Zhang, X., Ren, S., & Sun, J. (2016). Deep Residual Learning for Image Recognition. Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 770–778. https://doi.org/10.1109/CVPR.2016.90
Ibrahim, A., Hassan, M., & Youssef, K. (2024). High-Precision Organic Waste Classification using Fine-Tuned Object Detectors for Bio-Fertilizer Production. Waste Management, 178, 145–156. https://doi.org/10.1016/j.wasman.2024.02.010
Jocher, G., Nishimura, K., Contente, T., & Guo, R. (2022). Comparative Evaluation of YOLOv5 Architecture Variants for Industrial Real-Time Object Detection. Journal of Real-Time Image Processing, 19(4), 789–801. https://doi.org/10.1007/s11554-022-01230-0
Jocher, G., & others. (2020). YOLOv5. GitHub Repository. https://github.com/ultralytics/yolov5
Katadata Databox. (2024). Komposisi Sampah Nasional Berdasarkan Jenis.
Kaya, O., Yilmaz, M., & Aksoy, S. (2022). Inter-Class Misclassification Dynamics in Automated Organic and Inorganic Waste Sorting Systems. Waste Management, 142, 112–123. https://doi.org/10.1016/j.wasman.2022.02.015
Kementerian Lingkungan Hidup/Badan Pengendalian Lingkungan Hidup. (2025). KLH/BPLH Luncurkan Kebijakan Pemanfaatan Sampah Organik untuk Pupuk Organik, Dorong Swasembada Pangan Berkelanjutan. https://kemenlh.go.id/news/detail/klhbplh-luncurkan-kebijakan-pemanfaatan-sampah-organik-untuk-pupuk-organik-dorong-swasembada-pangan-berkelanjutan
Kim, S.-W., Park, J.-Y., & Lee, S.-H. (2023). Backbone Feature Retention and Fine-Tuning Convergence in Industrial Vision Inspection Systems. Computers in Industry, 145, 103820. https://doi.org/10.1016/j.compind.2022.103820
Kompas.com. (2024). Pupuk Organik Bersubsidi Akan Disalurkan Agustus 2024. https://money.kompas.com/read/2024/06/19/151200226/pupuk-organik-bersubsidi-akan-disalurkan-agustus-2024
Kumar, A., Sharma, R., & Patel, K. (2024). Mitigating False Positive Spikes and Background Contamination in Mid-Training Object Detection Regimes. Expert Systems with Applications, 248, 123450. https://doi.org/10.1016/j.eswa.2024.123450
Li, J., Wang, H., & Zhang, Q. (2022). Partial Layer Freezing Strategies for Accelerating Convergence in Deep Convolutional Neural Networks. IEEE Transactions on Neural Networks and Learning Systems, 33(8), 3840–3851. https://doi.org/10.1109/TNNLS.2021.3065201
Li, X., Zhang, W., & Zhao, J. (2022). Evaluating Parameter Adaptation and Feature Separation Dynamics in Intermediate Epochs of Deep Object Detectors. IEEE Transactions on Industrial Informatics, 18(11), 7540–7550. https://doi.org/10.1109/TII.2022.3168920
Liu, Y., Chen, X., Huang, Z., & Zhao, L. (2022). Performance Analysis of Standard YOLOv5 Architecture for Automatic Waste Classification Using Confusion Matrix Evaluation. Journal of Cleaner Production, 340, 130821. https://doi.org/10.1016/j.jclepro.2022.130821
Martinez, C., Gomez, A., & Fernandez, L. (2024). High-Precision Organic Waste Detection in Agricultural Automation via Deep Learning Approaches. Waste Management, 174, 201–212. https://doi.org/10.1016/j.wasman.2023.11.018
Medium. (2023). YOLO Algorithm Object Detection Visualization. https://medium.com/mlearning-ai/yolov5-object-detection-tutorial-3f82d2363b71
Mendoza, R., Santos, C., & Gomez, E. (2023). Computer Vision-Based Classification and Sorting of Organic Waste for Bio-Fertilizer Production. Waste Management, 163, 134–144. https://doi.org/10.1016/j.wasman.2023.03.015
Mittal, G., Yagnik, K., Garg, M., & Krishnan, N. (2016). Deep Learning Based Waste Segregation System. Proceedings of the IEEE Conference on Computer Vision Workshops, 1–6.
Nguyen, H.-T., Tran, D.-T., & Kim, H.-T. (2022). Suppressing Background False Positive Contamination in Deep Learning Vision Systems Using Data Augmentation. IEEE Access, 10, 89123–89135. https://doi.org/10.1109/ACCESS.2022.3199821
of Fuzzy Logic, J., & Authors, I. S. (2024). Features Exploitation of YOLOv5-Based Freeze Backbone for Performance Improvement of UAV Object Detection. International Journal of Fuzzy Logic and Intelligent Systems, 24(3), 194–202. https://doi.org/10.5391/IJFIS.2024.24.3.194
Padilla, R., Passos, W. L., Dias, T. L. B., Netto, S. L., & da Silva, E. A. B. (2021). A Survey on Performance Metrics for Object-Detection Algorithms. 9, 131061–131076. https://doi.org/10.1109/ACCESS.2021.3115262
Park, J.-W., Lee, S.-H., & Kim, S.-Y. (2023). Threshold Optimization and Convergence Analysis for Industrial Object Detection under Complex Backgrounds. Computers in Industry, 149, 103910. https://doi.org/10.1016/j.compind.2023.103910
Patel, K., Sharma, R., & Kumar, A. (2024). Enhanced YOLOv5 Architecture with Layer Freezing and Custom Augmentation under Complex Environmental Lighting. Expert Systems with Applications, 238, 122105. https://doi.org/10.1016/j.eswa.2023.122105
Pixabay. (2024). Various Types of Organic Waste and Garden Scraps. https://pixabay.com/photos/compost-organic-waste-garbage-1692742/
Prabowo, A., Rahayu, S., & Lestari, E. (2023). Computer Vision-Based Classification of Waste Materials for Bio-Fertilizer Feedstock Processing. Waste Management, 161, 102–111. https://doi.org/10.1016/j.wasman.2023.02.018
Prabowo, A., Rahayu, S., & Lestari, E. (2024a). Deep Learning-Driven Precision Sorting of Organic Feedstock for Bio-Organic Fertilizer Production. Waste Management, 176, 189–199. https://doi.org/10.1016/j.wasman.2024.01.015
Prabowo, A., Rahayu, S., & Lestari, E. (2024b). Optimization of Epoch Training and Inference Latency for Automated Waste Sorting in Petrochemical Industries. Computers and Electronics in Agriculture, 216, 108510. https://doi.org/10.1016/j.compag.2023.108510
Prasad, R., Verma, S., & Gupta, A. (2022). Identifying Optimal Training Duration and Stopping Criteria to Prevent Background Noise Activation in Deep Inspection Models. Pattern Recognition Letters, 160, 45–52. https://doi.org/10.1016/j.patrec.2022.06.008
PT Pupuk Indonesia (Persero). (2024). Penyaluran Pupuk Bersubsidi Tahun 2024.
Rahman, Md. A., Islam, Md. S., & Hossain, T. (2023). Mitigating Background False Positive Contamination in Deep Learning Object Detection for Industrial Sorting Lines. IEEE Access, 11, 45210–45221. https://doi.org/10.1109/ACCESS.2023.3271102
Rajaraman, S., Antani, S., & Jaeger, S. (2023). Analyzing Convergence Dynamics and Feature Adaptation in Intermediate Training Epochs for Deep Object Detectors. IEEE Access, 11, 18240–18252. https://doi.org/10.1109/ACCESS.2023.3245678
Redmon, J., Divvala, S., Girshick, R., & Farhadi, A. (2016). You Only Look Once: Unified, Real-Time Object Detection. ArXiv Preprint ArXiv:1506.02640. https://arxiv.org/abs/1506.02640
Researchers, S. I. (2026). Implementation of Organic and Inorganic Waste Detection Modeling in School Waste Bins Using YOLOv11. Journal of Esaprom. https://ejournal.seaninstitute.or.id/index.php/esaprom/article/view/7986
Santos, G., Oliveira, R., & Silva, F. (2023). Early-Epoch Training Dynamics and Misclassification Patterns in Real-Time Organic Waste Sorting. Waste Management, 160, 89–99. https://doi.org/10.1016/j.wasman.2023.01.012
Santosa, S. G., Kalsum, U., & Pribadi, E. M. (2023). Respon Pertumbuhan dan Produksi Kailan terhadap Perbedaan Penyiraman Otomatis dan Dosis Pupuk Organik Cair Limbah Pisang. Jurnal Pertanian Presisi, 7(2), 155–167. https://doi.org/10.35760/jpp.2023.v7i2.9942
Siddique, A., Hasan, M., & Islam, R. (2023). Neck and Head Weight Optimization Dynamics in Deep Object Detectors During Small-Epoch Regimes. Pattern Recognition Letters, 171, 55–63. https://doi.org/10.1016/j.patrec.2023.05.009
Su, C., Zhang, P., & Dong, H. (2023). Enhanced YOLO Architecture Optimization for Industrial Inspection in Complex Environmental Backgrounds. Expert Systems with Applications, 225, 120115. https://doi.org/10.1016/j.eswa.2023.120115
Szeliski, R. (2022). Computer Vision: Algorithms and Applications (2nd ed.). Springer Nature. https://doi.org/10.1007/978-3-030-34372-9
Tan, M., & Le, Q. V. (2019). EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks. Proceedings of the International Conference on Machine Learning (ICML), 6105–6114.
Tan, Z.-P., Wu, J.-H., & Zhou, C.-Y. (2023a). Convergence Stability and Misclassification Reduction in Extended Epoch Training of Deep Detectors. Computers and Electronics in Agriculture, 212, 108090. https://doi.org/10.1016/j.compag.2023.108090
Tan, Z.-P., Wu, J.-H., & Zhou, C.-Y. (2023b). Optimization of Inter-Class Feature Separation in Deep Convolutional Networks for Waste Sorting Applications. Computers and Electronics in Agriculture, 211, 107982. https://doi.org/10.1016/j.compag.2023.107982
Unsplash. (2024). Digital Concept of Computer Vision and Recognition. https://unsplash.com/photos/a-close-up-of-a-circuit-board-with-a-blue-light-L8tWZTmCcdo
Wang, C., & Zhang, J. (2023). Deep Learning-Based Organic and Inorganic Waste Sorting for Agricultural Bio-Fertilizer Production. Waste Management, 156, 112–121. https://doi.org/10.1016/j.wasman.2022.11.025
Wang, H., Liu, Y., & Sun, Q. (2023a). Baseline Performance and Inference Latency Analysis in Full-Weight Convolutional Object Detectors. IEEE Transactions on Industrial Informatics, 19(8), 8820–8830. https://doi.org/10.1109/TII.2022.3211050
Wang, H., Liu, Y., & Sun, Q. (2023b). Substantial Reduction of Background False Positive Contamination in Deep Object Detectors via Optimal Epoch Scaling. Expert Systems with Applications, 232, 120810. https://doi.org/10.1016/j.eswa.2023.120810
Wang, H., Liu, Y., & Sun, Q. (2024). Early Epoch Convergence Bottlenecks and Background False Positive Contamination in YOLO Architectures. Expert Systems with Applications, 245, 123110. https://doi.org/10.1016/j.eswa.2023.123110
Xu, Y., Liang, D., & Song, J. (2023a). Convergence Saturation and Over-fitting Dynamics in High-Epoch Fine-Tuning of YOLO Frameworks. Pattern Recognition, 142, 109680. https://doi.org/10.1016/j.patcog.2023.109680
Xu, Y., Liang, D., & Song, J. (2023b). Over-Sensitivity and Background Overfitting in Extended Epoch Training of YOLO Architectures. IEEE Transactions on Industrial Informatics, 19(9), 9512–9522. https://doi.org/10.1109/TII.2023.3241105
Xu, Y., Liang, D., & Song, J. (2023c). Over-Sensitivity and Background Overfitting Spikes in Extended Epoch Object Detection Regimes. Expert Systems with Applications, 234, 121015. https://doi.org/10.1016/j.eswa.2023.121015
Zhang, W., Liu, M., & Chen, X. (2023). Evaluating Convergence Saturation and Misclassification Dynamics in High-Epoch Object Detector Training. IEEE Transactions on Industrial Informatics, 19(10), 10210–10221. https://doi.org/10.1109/TII.2023.3254102
Zhang, W., Liu, M., & Chen, X. (2024). Analyzing and Mitigating Background False Positives in Object Detection Models under Small-Epoch Training Regimes. Pattern Recognition, 148, 110182. https://doi.org/10.1016/j.patcog.2023.110182
Zhang, Y., Chen, X., & Wang, L. (2021). Waste Detection Using Deep Learning: A Review. IEEE Access, 9, 12345–12360. https://doi.org/10.1109/ACCESS.2021.3050000
Zhao, R., Zhang, Y., Liu, B., & Wang, J. (2023). Extended Epoch Training and Convergence Optimization for Industrial Waste Classification Using YOLO Architectures. Computers and Electronics in Agriculture, 208, 107780. https://doi.org/10.1016/j.compag.2023.107780
Zhou, W., Yang, F., & Liu, B. (2023). Computer Vision-Based Classification of Organic Waste Materials for Bio-Fertilizer Production. Waste Management, 158, 78–88. https://doi.org/10.1016/j.wasman.2023.01.005
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Wirawan Setyo Prakoso, Alva Rischa Qhisthana Pratika, Ratih Kusuma Dewi, Luqman Aji Kusumo, Andi Amar Thahara
This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.
Authors who publish with this journal agree to the following terms:
1. Copyright Retention and Open Access License
Authors retain copyright of their work and grant the journal non-exclusive right of first publication under the Creative Commons Attribution 4.0 International License (CC BY 4.0).
This license allows unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
2. Rights Granted Under CC BY 4.0
Under this license, readers are free to:
- Share — copy and redistribute the material in any medium or format
- Adapt — remix, transform, and build upon the material for any purpose, including commercial use
- No additional restrictions — the licensor cannot revoke these freedoms as long as license terms are followed
3. Attribution Requirements
All uses must include:
- Proper citation of the original work
- Link to the Creative Commons license
- Indication if changes were made to the original work
- No suggestion that the licensor endorses the user or their use
4. Additional Distribution Rights
Authors may:
- Deposit the published version in institutional repositories
- Share through academic social networks
- Include in books, monographs, or other publications
- Post on personal or institutional websites
Requirement: All additional distributions must maintain the CC BY 4.0 license and proper attribution.
5. Self-Archiving and Pre-Print Sharing
Authors are encouraged to:
- Share pre-prints and post-prints online
- Deposit in subject-specific repositories (e.g., arXiv, bioRxiv)
- Engage in scholarly communication throughout the publication process
6. Open Access Commitment
This journal provides immediate open access to all content, supporting the global exchange of knowledge without financial, legal, or technical barriers.
