C2PA manifests and AI watermarks can independently validate contradictory claims on the same asset, and a cross-layer audit protocol resolves this with 100% accuracy on 3500 images.
Secured color image watermarking technique in dwt-dct domain
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
abstract
The multilayer secured DWT-DCT and YIQ color space based image watermarking technique with robustness and better correlation is presented here. The security levels are increased by using multiple pn sequences, Arnold scrambling, DWT domain, DCT domain and color space conversions. Peak signal to noise ratio and Normalized correlations are used as measurement metrics. The 512x512 sized color images with different histograms are used for testing and watermark of size 64x64 is embedded in HL region of DWT and 4x4 DCT is used. 'Haar' wavelet is used for decomposition and direct flexing factor is used. We got PSNR value is 63.9988 for flexing factor k=1 for Lena image and the maximum NC 0.9781 for flexing factor k=4 in Q color space. The comparative performance in Y, I and Q color space is presented. The technique is robust for different attacks like scaling, compression, rotation etc.
years
2026 2representative citing papers
AI content watermarking exhibits detection disparities across languages, cultures, and demographics due to content-dependent signal properties, with benchmarks failing to disaggregate performance and watermarking held to lower fairness standards than generative models.
citing papers explorer
-
Authenticated Contradictions from Desynchronized Provenance and Watermarking
C2PA manifests and AI watermarks can independently validate contradictory claims on the same asset, and a cross-layer audit protocol resolves this with 100% accuracy on 3500 images.
-
Who Gets Flagged? The Pluralistic Evaluation Gap in AI Content Watermarking
AI content watermarking exhibits detection disparities across languages, cultures, and demographics due to content-dependent signal properties, with benchmarks failing to disaggregate performance and watermarking held to lower fairness standards than generative models.