Wavelet-Domain Digital Image Watermarking Based on Human Visual System (HVS)

Wavelet-domain digital image watermarking technology based on the Human Visual System (HVS) involves multiple systematic steps. The process begins with watermark embedding, where watermark information is inserted into the wavelet domain of digital images using frequency masking properties of HVS to maximize invisibility. This is typically implemented through coefficient modification in selected subbands using algorithms like DWT (Discrete Wavelet Transform). Next, watermark extraction is performed to retrieve the embedded information from watermarked images, often utilizing correlation detection or threshold-based methods. The technology also requires image decomposition, where images are broken down into different frequency subbands (LL, LH, HL, HH) through multi-level wavelet transformation, creating optimal spaces for watermark embedding. Finally, image reconstruction is conducted to rebuild the processed images after watermark operations using inverse DWT, producing the final watermarked image while maintaining visual quality through HVS-adaptive quantization strategies.