A new type of art called mosaic is proposed here for automatically combining small tiles of secret image to form a target in the sense of mosaic. To create the mosaic image, we first find the similar target image for selected source image. A novel algorithm technique is proposed here to ﬁnd a similar tile image in the secret image to ﬁt into each block in the target image. The information of placing the tile image fitting sequence in target image is embedding into random selected pixel in created mosaic image. We can’t recover the image without the fitting sequence.
One of the method to create crystallization mosaic images using voronoi diagrams by placing blocks at random sites and filling colors into the blocks based on the content of the original image. The target image selection is done by fetches the image with the most similar content from an image database. The 3D image is convert into 1D image for creating the mosaic. Secret image tiles are placed in target image according to the information presented in key file.
While converting 3D image to 1D the original color value of image are get dropped. If the database doesn’t contain similar image for the selected secret image, then itself assign a low similar image as target image. The visibility of target image is get more blind by placing tile in target image itself.
The 3D image is considered as it as 3D image for creating mosaic image. We don’t use database for selecting the target image for given source image. We select the target image by Selecting manually. Tile images are placed in blank image created by the user instead of placing in target image. Instead of filling the color of secret, we place the tile into the created blank image by user.
The image quality is high comparing to the existing system, by placing image into blank created image. Tile image placing are compared with the entire target image and choose the perfect match for the tile. The original image is gotten finally, because of using the 3D as it as 3D image. Here there is no conversion process is appear.
Similar Target Image: Selecting the secret image for creating mosaic image, after that picking the most similar target image for the selected source image. We have an efficient image similarity algorithm for comparing similarity value between both the images. The value returned here display’s the how the match the both the images are.
Image Splitting : Then we have create the tile images for creating a mosaic image, for that we want to split the source image into particular size small image called as tile. The tile images are compared with the target image to getting the position for placing the tile into the target image.
Mosaic Creation: Then we have to find the position which is perfectly matching for each every tile of secret image. After finding the position of each tile, then we have to create a new blank image for placing the tile in it. Then placing the tile images based on the information presented in the fitting sequence. Then we can generate the duplicate tile images for completing the image as target image.
Fitting Sequence Embedding: Then embedding the fitting sequence file into the created mosaic image by using an secret novel algorithm to hide the fitting sequence information . By using this fitting sequence the receiver can able to reconstruct the original image as it send by the sender. The fitting sequence containing the information about the image name and where it is perfectly matched in target image position.
Reconstruct Secret Image: After receiving the mosaic image, we can first de-embedding the fitting sequence file, because with help of this only we split the tile images presented in the mosaic image. Then we have to split the mosaic created image based on the information presented in fitting sequence and named it as, the name itself presented in file. Then finally compared the images got from the mosaic image, we got the image that sender can send.
Software Requirement: Jdk1.6.0_07 Netbeans IDEHardware Requirement: PENTIUM IV-2.7 GHZ 1 GB DDR RAM 250 GB HARD DISK
The new art which can be used for secure keeping or covert communication of secret images. This type of mosaic image is composed of small fragments of an input secret image and though all the fragments of the secret image can be seen clearly, they are so tiny in size and so random in position that people cannot ﬁgure out what the source secret image looks like. A novelalgorithm has also been proposed for searching the tile images in a secret image for the most similar ones to ﬁt the target blocks of a selected target image more efﬁciently.
This is applicable only for colored images, our future work is to work for grayscale images also. Then next one is reducing the time consumption for creating the mosaic image.
H. Narasimhan and S. Satheesh, “A randomized iterative improvement algorithm for photomosaic generation,” in Proc. NaBIC, Coimbatore, India, Dec. 2009, pp. 777–781. S. Battiato and G. Puglisi, “3D ancient mosaics,” in Proc. ACM Int. Conf. Multimedia, Florence, Italy, Oct. 2010, pp. 1751–1753. D. Coltuc and J. M. Chassery, “Very fast watermarking by reversible contrast mapping,” IEEE Signal Process. Lett.,vol.14,no.4,pp. 255–258, Apr. 2007 Y.-S.Choi,B.-k.Koo,andB.-R.Moon,“Optimizationofanimage set by genetic feature selection for real-time photomosaics,” in Proc. GECCO, Portland, OR, Jul. 2010, pp. 1309–1310. S. Battiato, C. Guarnera,G.DiBlasi,G.Gallo,andG.Puglisi,M. Bubak, Ed. et al., “A novelartiﬁcial mosaic generation technique driven by local gradient analysis,” in Proc. ICCS, Crakov, Poland, Jun. 2008, vol. 5102, pp. 76–85
need screen shots of corresponding project, please mail me….
For any IEEE project needs contact me,ARUL Jarulj7978@hotmail.com+91 9944797816