What are the techniques of spatial domain watermarking?

The spatial domain define the image in the form of pixels. The spatial domain watermarking embeds the watermark by changing the intensity and the colour value of some preferred pixels.

The spatial domain watermarking is simpler and its calculating speed is high than transform domain but it is less powerful against attacks. The spatial domain techniques can be simply used to any image.

There are various techniques of spatial domain watermarking which are as follows −

Least Significant Bit (LSB) − The LSB is the easiest spatial domain watermarking technique to insert a watermark in the least significant bits of some randomly selected pixels of the cover image.

The main advantage of this method is that it can be simply performed on images and it supports high perceptual transparency. When it can embed the watermark by utilizing LSB the quality of the image will not degenerate. 

The main disadvantage of LSB technique is its poor robustness to common signal processing services because by using this approach watermark can simply be damaged by any signal processing attacks. It is not vulnerable to attacks and noise but it is very much undetectable.

Additive Watermarking − The basic approach for embedding the watermark in spatial rule is to insert pseudo random noise design to the strength of image pixels.

The noise signal is generally integers such as (-1, 0, 1) or sometimes floating point numbers. It can provide that the watermark can be detected, the noise is produced by a key, such that the correlation among the numbers of multiple keys will be very low.

SSM Modulation Based Technique − Spread-spectrum techniques are approaches in which energy created at several discrete frequencies is intentionally development or delivered in time.

SSM based watermarking algorithms embed data by linearly joining the host image with a limited pseudo noise signal that is modulated by the embedded watermark.

Texture mapping coding Technique − This method is beneficial in only those images which have some texture element in it. This method conceal the watermark in the texture element of the image.

This algorithm is only applicable for those areas with high number of arbitrary texture images and cannot be completed automatically. This method hides information within the continuous random texture designs of an image.

Patchwork Algorithm − Patchwork is a data hiding approach produced by Bender et al and published on IBM Systems Journal, 1996. It depends on a pseudorandom, statistical model. Patchwork imperceptibly adds a watermark with a specific statistic utilizing a Gaussian distribution.