For Project Enquiry +91 9962 067 067

EXISTING SYSTEM

PROPOSED SYSTEM

EXISTING CONCEPT:

          We created a challenging real-world copy-move dataset, and a software framework for systematic image manipulation.

         Experiments show, that the features SIFT and SURF, as well as the block-based DCT, DWT, KPCA, PCA and ZERNIKE features perform very well.

PROPOSED CONCEPT:

          We propose a new framework for CMFD; the test image is first segmented into non-overlapped patches.

          The aim of the first stage is to find the suspicious matches, and a transform matrix between them is roughly estimated. Then in the second stage we confirm the existence of CMF by means of refining the transform matrix.

EXISTING  TECHNIQUE :

          SIFT ALGORITHMS

PROPOSED ALGORITHM:

          Key point Extraction and Description

          Matching Between Patches

TECHNIQUE DEFINITION:

Unlike block-based algorithms, SIFT methods rely on the identification and selection of high-entropy image regions. A feature vector is then extracted per key point.

ALGORITHM DEFINITION:

          In the computer science field of artificial intelligence, genetic algorithm (GA) is a search heuristic that mimics the process of natural selection.

DRAWBACKS:

          It takes more time consumption.

          As presented, the copied regions are meaningful, i. e. either they hide image content, or they emphasize an element of the picture.           The software allows the snippets to be inserted at arbitrary positions.

ADVANTAGES:

          It takes less time than existing system.

          The block-based methods usually need a huge amount of time to detect an image.

          In this regard, the key point-based methods are faster and more favorable than the block-based ones, because the number of the image key points is smaller than that of the divided blocks.