Augmented Reality means interacting with the real world circumstances and virtual objects at the same time which is augmented by system generating inputs. That is, it allows the system generated content to be superimposed over a live camera of the real world. It is an absolutely graphical experience that is combined with real world circumstances. Here, the landmark will be acquired as an input of the image by sight-based camera parameter estimation methods by using SFM (STRUCTURE FROM MOTION) technique. In this paper, we are going to overlay real still images of landmark with virtual computer graphical images. These SFM techniques which have been discussed earlier in some of the previous papers will not work in real time applications because of its high cost. But here, this method will achieve fast and accurate landmark estimation based on SFM and we obtain the 3-D structure of a landmark by using this technique which has no need of information about the scene.

Image encryption is used to protect the data of an image in the most secure way. In this paper, we perform encryption and decryption of the three color planes based on gray code effect and fast Fourier transform. Gray level codes indicate that occurrence of one bit changes from a previous state of current. This process allows the system to detect the error when more than one bit changes occur. It is more efficient than binary bit plane decomposition. The Fourier transform is mainly used to represent the frequency characteristic of a spatial domain image. The color plane scrambling process is done by combining three individual encrypted planes into one composite image. This method which is applicable for any type of image formats as bmp, JPEG, GIF. This paper gives a simple way to scramble the image in unpredictable way. It provides more security than other encryption algorithm. The original images can be retrieved using a correct security key which achieves better results than the other existing methods. By exploiting gray code and Fourier transforms it increases the difficulty of decoding.