Home // SENSORCOMM 2025, The Nineteenth International Conference on Sensor Technologies and Applications // View article

Invisible Watermarking for Image Data Protection in Sensor Network Environments

Authors:
Seo-Yi Kim
Na-Eun Park
Il-Gu Lee

Keywords: Sensor camera; Digital watermarking; Image protection.

Abstract:
Advancements in Artificial Intelligence (AI) have greatly increased the risk of digital-image tampering, underscoring the need to verify the integrity and authenticity of image data collected and transmitted within sensor networks and sensor-based systems. As visual threats, such as deepfakes and adversarial attacks proliferate, manipulated sensor images can trigger severe security incidents and false detections. This paper proposes a robust watermarking method that employs a three-level Discrete Wavelet Transform (DWT) and Singular Value Decomposition (SVD) to repeatedly embed a watermark into the singular values of both low- and selected high-frequency components. Designed to account for transmission noise and environmental distortions in multi-sensor settings, the proposed approach leverages redundancy across multiple frequency bands to enhance resistance to diverse signal-distortion attacks while keeping the watermark imperceptible. Experimental results show that the proposed method significantly surpasses conventional techniques in watermark extraction accuracy while preserving high image quality, establishing it as a reliable security solution for protecting image integrity and detecting tampering in sensor-based environments.

Pages: 26 to 33

Copyright: Copyright (c) IARIA, 2025

Publication date: October 26, 2025

Published in: conference

ISSN: 2308-4405

ISBN: 978-1-68558-304-0

Location: Barcelona, Spain

Dates: from October 26, 2025 to October 30, 2025