Steganography is the technique for hiding secret information in any ordinary image. Such image in which secret information is hidden is known as stego image. Where as Steganalysis is the reverse technique in which detection of the secret information is done in the stego images .In this paper, various visual and statistical attack techniques are implemented for the detection of the secret data in the stego image.

There are two ways of doing visual attack: 1. The technique in which detection is done by knowing the cover image without the knowledge of the coding algorithm for the formation of stego image. This technique can be named as visual attack: known cover image. 2. Another technique in which detection is applied to the incoming image without any information known as universal attack .This technique can be named as visual attack: Enhanced LSB technique. With these two techniques the area of hidden data is detected on the basis of the difference in visual texture of the stego image.

Similarly two ways of doing statistical attack: 1. The technique in which detection is done by comparing the stego & cover image in which first order statistical analysis is applied without the knowledge of the coding algorithm of the stego image. This technique can be named as histogram analysis. 2. Another technique in which detection is done by the image smoothening technique by using probability distribution function. This technique can be named as statistical attack: image smoothening. With these two technique based on the difference in statistical parameter of the stego image with cover image detection is done. Further the comparison of the techniques of both the attacks is done on the basis of time, memory requirement and information requirement.