EUVIMEDEUVIMED entdecken
Crossref

AI-Based Intrusion Prediction System Using YOLOv8 and Behavioral Movement Analysis

Akshay V B, Varsha C R

International Journal of Sciences and Innovation Engineering · 12.03.2026 · 3 (3) · 417-421

DOI: 10.70849/ijsci03032600536 ↗

Kurz gefasst

Fachartikel in der Zeitschrift International Journal of Sciences and Innovation Engineering, online erschienen am 12.03.2026. Verfasst von Akshay V B und 1 weiteren Autor:innen.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Security surveillance systems play an important role in protecting restricted environments such as industrial facilities, military zones, and other sensitive infrastructures. Traditional surveillance systems rely heavily on continuous human monitoring of CCTV feeds, which can result in delayed responses, operator fatigue, and increased chances of missing critical security events. Most automated systems focus only on detecting intrusions after a restricted boundary has already been crossed, limiting the ability to respond proactively. To address these limitations, this research proposes an AI-Based Intrusion Prediction and Detection System that emphasizes early identification of potential intrusion events through behavioral analysis. The system uses the YOLOv8 object detection model to detect and track individuals in video frames, while a polygon-based boundary defines restricted zones within the surveillance area. Instead of waiting for an actual boundary violation, the system continuously analyzes factors such as direction of movement, trajectory consistency, and approach behavior toward the restricted boundary consistency to estimate the likelihood of intrusion before it occurs. A dynamic risk value is generated and updated in real time as individuals approach the restricted region. When the predicted risk reaches a critical level or when the boundary is crossed, the system confirms the intrusion event, captures a snapshot as evidence, and sends an automated email alert to the authorized user. This predictive surveillance approach improves response time, enhances monitoring efficiency, and significantly reduces dependence on continuous manual observation while maintaining real-time system performance.

Zitieren

APA

B, A. V., & R, V. C. (2026). AI-Based Intrusion Prediction System Using YOLOv8 and Behavioral Movement Analysis. International Journal of Sciences and Innovation Engineering, 3(3), 417-421. https://doi.org/10.70849/ijsci03032600536

Vancouver

B AV, R VC. AI-Based Intrusion Prediction System Using YOLOv8 and Behavioral Movement Analysis. International Journal of Sciences and Innovation Engineering. 2026;3(3):417-421. doi:10.70849/ijsci03032600536

BibTeX
@article{B2026,
  author = {B, Akshay V and R, Varsha C},
  title = {AI-Based Intrusion Prediction System Using YOLOv8 and Behavioral Movement Analysis},
  journal = {International Journal of Sciences and Innovation Engineering},
  year = {2026},
  volume = {3},
  number = {3},
  pages = {417--421},
  doi = {10.70849/ijsci03032600536}
}
Kostenlos starten →