Text fusion in images is an important technology for image processing. We have lots of important information related to the patient's reports and need lots of space to store and the proper position and name which relates that image with that data. In our work we are going to find out the ROI (region of interest) for the particular image and will fuse the related document in the NROI (non-region of interest) of the image, till yet we have many techniques to fuse text data in the medical images one of form them is to fuse data at the boarders of the images and build the particular and pre-defined boarder space. We have propose an algorithm in which we first find out the area of interest and after that we find noisy pixels of the image to embed data in that noisy portions of the image. We use wavelets for smoothing images and segmentation process for extracting region of interest. Coordinates of the noisy pixels have been located and data has been embedded in those pixels .The used embedding technique embed data in least significant bits, hence does not degrade the quality of the image to unacceptable limits. Results show that it gives good PSNR and MSE values which are used for measuring quality effected performance.
展开▼