.png){kind=icon}
.png){kind=icon}
Return
Pixel level analysis to detect and mitigate tampering in surveillance video
.png){kind=icon}
PRE
AI
J
R
DOI：10.1007/s11042-026-21852-z.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Video tampering detection in surveillance footage poses major challenges for forensic applications, as subtle tampering can mask authentic content while preserving visual plausibility, thereby compromising evidential integrity. In this work, a PECURE (POB-ECC Inpaint-based Unified Restoration) mechanism is proposed for identifying regions of interest (ROIs) by analyzing intensity variations in the HSV color space. The error correction using Error Convolutional Codes (ECC) and Permutation Ordered Binary (POB) repairs corrupted data, while inpainting refines and reconstructs the affected areas, ensuring optimal restoration quality. The restored regions show a peak signal-to-noise ratio (PSNR) of 32.5 dB and a structural similarity index measure (SSIM) of 0.92, implying high reconstruction precision and minimal perceptual distortion. The distortion noise and mean squared visual salience errors were around 0.003, resulting in significant improvements in image quality and fidelity. This confirms the effectiveness of PECURE to preserve original image details and prevent alterations, thus ensuring data integrity and authenticity.
Keywords:
Error convolutional code (ECC)
Permutation ordered binary (POB)
Digital forensics
Inpainting
Tampering detection
Video restoration
Journal
IF:
3
Papers:
1.9W
Citations:
3.2W
Organization
Cited Papers