Digital Watermark Padding for Moving Image Noise Tolerance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current digital watermarking techniques fail to detect watermarks in real-time from moving images, such as those on a TV screen, due to noise issues during image capture and difficulty in identifying the watermark detection subject within captured frames, especially under varying camera angles and backgrounds.
Innovation Solution
A digital watermark embedding method that involves inputting moving image data with watermark information and pattern switching information, generating and superimposing temporal changes in watermark patterns onto each frame image, and reconstructing moving image data for reliable detection, along with a detection method that uses difference images to identify and extract watermark patterns from captured frames.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If digital watermark detection is performed on moving images captured by camera, then real-time information service can be realized, but noise tolerance is insufficient due to D/A and A/D conversion
Solution Approach 1:
The patent embeds synchronization information and watermark patterns into the moving image data before capture. The watermark patterns are pre-modulated with synchronization signals that enable the detection device to identify and track the watermark even in noisy captured images, allowing real-time detection while maintaining noise tolerance through the preliminary embedding of reference information.
Solution Approach 2:
The patent employs feedback mechanisms where the detection device continuously compares captured frame images with reference images containing embedded synchronization information. This feedback loop enables real-time adjustment and compensation for noise introduced during D/A and A/D conversion, maintaining detection reliability while achieving real-time processing capability.
2Productivity
If digital watermark detection is performed on captured frame images, then real-time detection is enabled, but it is difficult to identify the detection subject in varying conditions
Solution Approach 1:
The patent uses distinct synchronization patterns and watermark signatures that create detectable visual differences between the watermark regions and background images. These synchronized patterns act as visual markers that stand out under various camera angles and lighting conditions, enabling easy identification of the detection subject in real-time captured images.
Solution Approach 2:
Synchronization information and distinctive watermark patterns are embedded into the moving image data before capture. These pre-placed reference markers serve as identification cues that enable the detection device to quickly locate and identify the detection subject in captured frames, regardless of camera angle or background variations.
3Reliability
If watermark patterns are embedded in moving images, then detection reliability is improved, but the complexity of embedding and detection processes increases
Solution Approach 1:
The patent divides the moving image data into sequential frame images and processes them individually. The watermark embedding is performed on each frame separately with synchronized patterns, and detection is conducted frame-by-frame. This segmentation approach improves detection reliability through systematic processing while reducing overall system complexity by breaking down the complex task into manageable, standardized operations.
Data Source
AI summary
A digital watermark embedding method of the present invention includes: a step of sequentially obtaining each frame image of the moving image data and frame display time; a step of generating a watermark pattern using watermark information, the frame display time and watermark pattern switching information; a step of superimposing the watermark pattern onto the frame image, and combining watermark embedded frame images obtained by sequentially repeating the processes to generate watermark embedded moving image data. A digital watermark detection method includes a step of sequentially obtaining a frame image; a step of generating a difference image between the currently obtained frame image and a previously obtained frame image; and a step of performing digital watermark detection from the difference image to output digital watermark detection status, and when digital watermark detection process is continued, obtaining a new frame again to repeat the above processes.