PERANCANGAN MODEL YOLOv8-LSTM UNTUK DETEKSI GERAKAN ANOMALI
Abstract
Manual monitoring of CCTV systems for detecting anomalous movements, such as criminal activity, is highly inefficient and prone to human error, thus urging the need for automated surveillance systems. A key research gap is that most object detection models (spatial in nature) fail to understand the temporal context (movement patterns over time) which is key to distinguishing normal and anomalous activities. This study proposes the design of a hybrid deep learning model YOLOv8-LSTM to address this issue. Using the 4D R&D (Define, Design, Develop) research methodology, an architecture is designed in which YOLOv8 (yolov8m) functions as a spatial feature extractor (generating a 106-dimensional vector) from each video frame. The sequence of these features is then analyzed using a Bidirectional Long Short- Term Memory (Bi-LSTM) equipped with an Attention Pooling mechanism to model temporal dependencies and classify movements. The prototype test results on the test set show strong performance, achieving an AUC of 0.8646 and an F1-Score of 0.6530. Qualitative analysis through 3D latent space visualization successfully demonstrated the model's effectiveness: initially overlapping spatial features (YOLOv8 input) were successfully mapped into clearly separated clusters of normal and anomalous classes (LSTM output). This study validates that the proposed hybrid architecture effectively combines spatial and temporal understanding for accurate anomalous motion detection.
Downloads
References
Al-slemani, A. S. A., & Zengin, A. (2023). A New Surveillance and Security Alert System Based on Real-Time Motion Detection. Journal of Smart Systems Research, 4(1), 31–47. https://doi.org/10.58769/joinssr.1262853
Ashby, M. P. J. (2017). The Value of CCTV Surveillance Cameras as an Investigative Tool: An Empirical Analysis. European Journal on Criminal Policy and Research, 23(3), 441–459. https://doi.org/10.1007/s10610-017-9341-6
Chang, C.-W., Chang, C.-Y., & Lin, Y.-Y. (2022). A hybrid CNN and LSTM-based deep learning model for abnormal behavior detection. Multimedia Tools and Applications, 81(9), 11825–11843.
Diwan, T., Anirudh, G., & Tembhurne, J. V. (2023). Object detection using YOLO: challenges, architectural successors, datasets and applications. Multimedia Tools and Applications, 82(6), 9243–9275. https://doi.org/10.1007/s11042-022-13644-y
Fahlevvi, M. R., Kusuma, K. A. P. I., & Anugerah, M. W. (2025). Integrasi Teknologi Digital dalam Pengawasan Internal Inspektorat Daerah Kabupaten Gianyar. JURNAL SYNTAX IMPERATIF : Jurnal Ilmu Sosial Dan Pendidikan, 6(2), 236–249. https://doi.org/10.54543/syntaximperatif.v6i2.687
Ganagavalli, K., & Santhi, V. (2024). YOLO-based anomaly activity detection system for human behavior analysis and crime mitigation. Signal, Image and Video Processing, 18(Suppl 1), 417–427.
Guan, Z. (2023). Real time object recognition based on YOLO model. Theoretical and Natural Science, 28(1), 137–143. https://doi.org/10.54254/2753-8818/28/20230450
Jocher, G., Chaurasia, A., & Qiu, J. (2023). Ultralytics YOLOv8. https://github.com/ultralytics/ultralytics
Khayrat, A., Malak, P., Victor, M., Ahmed, S., Metawie, H., Saber, V., & Elshalakani, M. (2022). An intelligent surveillance system for detecting abnormal behaviors on campus using YOLO and CNN-LSTM networks. 2022 2nd International Mobile, Intelligent, and Ubiquitous Computing Conference (MIUCC), 104–109.
KULA, S. (2015). The Effectiveness of CCTV in Public Places: Fear of Crime and Perceived Safety of Citizens. Bartin University Journal of Faculty of Economics & Administrative Sciences/Bartn Üniversitesi .Iktisadi ve .Idari Bilimler Fakültesi Dergisi, 6(12).
Lisdiana, L., Trihandayani, T., & Novaria, E. (2025). ANALISIS PENGAWASAN DIGITAL MELALUI CLOSED CIRCUIT TELEVISION DALAM PENCEGAHAN TINDAK PIDANA KRIMINALITAS DI WILAYAH KERJA KEPOLISIAN SEKTOR KECAMATAN ILIR TIMUR II KOTA PALEMBANG. CENDEKIA: Jurnal Ilmu Pengetahuan, 5(2), 737–746. https://doi.org/10.51878/cendekia.v5i2.5101
Liu, Y., Liu, J., Zhao, M., Yang, D., Zhu, X., & Song, L. (2022). Learning Appearance-Motion Normality for Video Anomaly Detection. 2022 IEEE International Conference on Multimedia and Expo (ICME), 1–6. https://doi.org/10.1109/ICME52920.2022.9859727
Nguyen, T.-N., & Meunier, J. (2019, October). Anomaly Detection in Video Sequence With Appearance-Motion Correspondence. Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV).
Sherstinsky, A. (2020). Fundamentals of Recurrent Neural Network (RNN) and Long Short-Term Memory (LSTM) network. Physica D: Nonlinear Phenomena, 404, 132306. https://doi.org/https://doi.org/10.1016/j.physd.2019.132306
Sohan, M., Sai Ram, T., & Rami Reddy, Ch. V. (2024). A Review on YOLOv8 and Its Advancements. In I. J. Jacob, S. Piramuthu, & P. Falkowski-Gilski (Eds.), Data Intelligence and Cognitive Informatics (pp. 529–545). Springer Nature Singapore.
Tay Nian Chi and Connie, T. and O. T. S. and G. K. O. M. and T. P. S. (2019). A Robust Abnormal Behavior Detection Method Using Convolutional Neural Network. In Y. and I. A. A. A. and A. P. Alfred Rayner and Lim (Ed.), Computational Science and Technology (pp. 37–47). Springer Singapore.
V, V., K, C. R., & C., R. A. (2022). Real Time Object Detection System with YOLO and CNN Models: A Review. https://arxiv.org/abs/2208.00773
Varghese, R., & M, S. (2024). YOLOv8: A Novel Object Detection Algorithm with Enhanced Performance and Robustness. 2024 International Conference on Advances in Data Engineering and Intelligent Computing Systems (ADICS), 1–6. https://doi.org/10.1109/ADICS58448.2024.10533619
Waruwu, M. (2024). Metode Penelitian dan Pengembangan (R&D): Konsep, Jenis, Tahapan dan Kelebihan. Jurnal Ilmiah Profesi Pendidikan, 9(2), 1220–1230. https://doi.org/10.29303/jipp.v9i2.2141
Welsh, B. C., & Farrington, D. P. (2009). Public Area CCTV and Crime Prevention: An Updated Systematic Review and Meta‐Analysis. Justice Quarterly, 26(4), 716–745. https://doi.org/10.1080/07418820802506206
Zou, Y. (2025). Research on Laboratory Abnormal Behavior Recognition Model for Security Management Based on YOLOv5 and LSTM. 2025 3rd International Conference on Data Science and Network Security (ICDSNS), 1–7. https://doi.org/10.1109/ICDSNS65743.2025.11168576
Copyright (c) 2026 Frolentika Frolentika, Hebert Karsten Juwono, Luigi Emiliandra, Kevin Pierre Rafael Sabran, Windy Sulistiawati, Rahmi Yulia Ningsih
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
Jurnal allows anyone to compose, correct, and do derivative works, even for commercial purposes, as long as they credit for the original work. This license is the freest. It is recommended for maximum distribution and use of licensed material.
The submitted paper is assumed not to contain any proprietary materials that are not protected by patent rights or patent applications; The responsibility for technical content and protection of proprietary materials rests with the authors and their organizations and not the responsibility of journal or its editorial staff. The primary (first/appropriate) author is responsible for ensuring that the article has been viewed and approved by all other authors. The author's responsibility is to obtain all necessary copyright waivers to use any copyrighted material in the manuscript before submission.
Jurnal Pendidikan, Sains dan Teknologi allows the author(s) to hold the copyright without restrictions and allow the author(s) to retain publishing rights without restrictions. Jurnal Pendidikan, Sains dan Teknologi CC-BY-SA or an equivalent license as the optimal license for the publication, distribution, use, and reuse of scholarly work. Jurnal Pendidikan, Sains dan Teknologi allows the author(s) to hold the copyright without restrictions and allow the author(s) to retain publishing rights without restrictions. Jurnal Pendidikan, Sains dan Teknologi CC-BY-SA or an equivalent license as the optimal license for the publication, distribution, use, and reuse of scholarly work.
In developing strategy and setting priorities Jurnal Pendidikan, Sains dan Teknologi recognize that free access is better than priced access, libre access is better than free access, and libre under CC-BY-SA or the equivalent is better than libre under more restrictive open licenses. We should achieve what we can when we can. We should not delay achieving free in order to achieve libre, and we should not stop with free when we can achieve libre.
Jurnal Pendidikan, Sains dan Teknologi is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
You are free to:
- Share a copy and redistribute the material in any medium or format
- Adapt a remix, transform, and build upon the material for any purpose, even commercially.
- The licensor cannot revoke these freedoms as long as you follow the license terms.
