Digital Image Processing

M. Jayapriyan ¹, J. S. Leena Jasmine ²

Affiliations
1 Velammal Engineering College, Chennai, TN, IN
2 Department of Computer Science, PU, IN

Abstract

A new reversible watermarking has been proposed, which involves histogram shifting modulation and takes care of the local specificities of the image content. These methods allows the user to restore exactly the original image from its watermarked version by removing the watermark and thus it becomes possible to update the watermark content with less distortion. In this bidirectional histogram shifting modulation both the payload and PSNR value will be high were we can embed the secret image with less distortion and error. By applying it to the image prediction-errors and by proposed inserts data in the textured areas, a classification process is used for identifying parts of the image that can be watermarked with most suited reversible modulation. This classification is based on a reference image derived from the image itself, a prediction of it which has the property of being invariant to watermark insertion. In that way the watermark embedder and the extractor remain synchronized for image extraction and reconstruction. Here we are going to use Odd-Even Histogram shifting method through which we can hide two images for watermarking and based on the type application the data may vary for example the data can be an image, text or binary values. The experiments conducted so far on some natural images and on medical images show that capacities smaller than 0.7 bpp here we can insert more data with lower distortion. In Bidirectional Histogram Shifting we are going to embed and extract two or more images were the PSNR value and the Payload will be high while comparing to the existing watermark.

Keywords

Reversible Watermarking, Medical/Military Image, Data Hiding, Predictive Errors, PSNR.

Full Text

     

P. Mathivanan ¹, M. Shrikalaa ², J. S. Leena Jasmine ³

Affiliations
1 Velammal Engineering College, IN
2 Arunai Engineering College, IN
3 Velammal Engineering College, Chennai, IN

Abstract

A novel steganography method with maximum data hiding capacity using point feature matching is performed in our proposed method. In order to increase data hiding, 2D Anaglyphic stereography images are used for this process. The 2D Anaglyphic stereo input images of different colours and coloured filters in front of each eye allows only the appropriate image to pass. Some international standards body has decided that the right eye filter should be blue, while the left eye filter is red, but this can vary. In our process these input images are obtained by a simple process, here adjacent frames from an input video is taken. The individually obtained adjacent frames act similar to stereo inputs, further these input images are subjected to preprocessing for better stegano process are individually steganoed with a secret data (image). The data hiding process is obtained by a simple method of stegano algorithm called LSB (Least Significant bit), here LSB bit of our cover image is replaced by a secret data. Later this individual stereo images with secret data is applied to Stabilization Using Point Feature Matching, this stabilization algorithm involves two steps. First, we determine the affine image transformations between all neighbouring frames of a video sequence using a Random Sampling and Consensus (RANSAC) procedure applied to point correspondences between two images. Second, we warp the embedded 2D Anaglyphic steganoed images to achieve a stabilized 3D stereo image with high data hiding capability, which contains two secret data in a single 3D stereo image.

Keywords

2D Anaglyphic Stereo Images, LSB, Random Sampling and Consensus (RANSAC).

Full Text