In this paper, a perceptual encryption algorithm is proposed. The perceptual feature requires that the quality of visual data be partially degraded by encryption. This algorithm selectively encrypts fixed-length code words in video input with three control factors. It provides useful features, viz. strict-size preservation, on-the-fly encryption and multiple perceptibilities. This makes it possible to support applications with different requirements. In addition, this algorithm is secure against ciphertext only attack and Known/Chosen -plaintext attack.Keywords -ciphertext-only attack, fixed length code word, perceptual encryption, secret key.

The proposed system in this paper gives the invisible watermarking which is performed by using Discrete Wavelet Transform. To get the invisible watermarking the alternate pixel value of the host video is replaced by the pixel value of watermark video/image. This type of watermarking provides a means of forensic analysis for combating media piracy. Video watermarking provides robustness to geometric attack such as rotation, cropping, contract altercation, time editing without compromising the security of the watermark.

Over the last couple of years several encryption algorithms applied to MPEG video streams emerged. Most of the algorithms tried to optimize the encryption process with respect to encryption speed. In this paper we will describe, evaluate and compare four MPEG encryption algorithms. Video Encryption Algorithm (VEA),Modified video encryption algorithm (MVEA), Real time Video Encryption Algorithm (RVEA), Perceptual Video Encryption Algorithm (PVEA) with respect to their encryption speed, security level and string size.PVEA is a generalized version of VEA for Perceptual Encryption, by selectively encrypting FLC data element in the video stream. Encrypting FLC data element makes PVEA maintain the strict-size preservation feature.