Data encryption is widely used to ensure security in open networks such as the internet. Most of the available encryption algorithms are used for text data; however, encryption of images is different from that of texts due to some intrinsic features of images such as bulk data capacity and high redundancy, which is generally difficult to handle by traditional methods. In most of the natural images the values of the neighboring pixels are strongly correlated. This means that the value of any given pixel can be reasonably predicted from the values of its neighbors. Basics of image encryption are explained and some popular techniques are also discussed and compared. Meanwhile a robust encryption scheme is also designed based on the combination of image permutation and encryption using orthonormal basis vectors. This allows the encrypted objects to be protected with high levels of security. Complete design process is explained and implemented step by step using MATLAB® 2010. The principle behind the presented solution may be applied to various image, audio, and video systems. The experimental results verify that the presented algorithm is a good approach for multimedia encryption.