T he applications accessing multimedia systems and content over the Web have grown immensely in the past five years. Furthermore, many end users can easily use tools to synthesize and edit multi- media information. Thus, security has become one of the most significant problems for distributing new information technology. It is necessary to pre- vent illegal copying, misappropriation, and mis- representation of digital audio, images, and video because they can be so easily copied and multi- plied without information loss. It’s also important to determine where and how much a multimedia file differs from its original. Thus, a need exists for developing technology that will help protect the integrity of digital content and secure the intellec- tual-property rights of owners. Watermarking is becoming the key method for protecting digital elements such as image, video, and sound. Digital watermarking embeds a signal into the original element, and the signal uniquely identifies the owner. This requires security solu- tions for such fields as distributed production processes and e-commerce because the producers seek to provide access control mechanisms to pre- vent their material’s misuse and theft. Watermarking development To better manage digital content security, researchers have evolved watermark processing in three categories according to specific applications’ requirements: robust, fragile, and semifragile watermarks. Robust watermarking resist attempts to remove or destroy the watermark. Primary applications are copyright protection and content tracking. Fragile watermarks can be easily destroyed. Authentication applications use such kinds of watermarks. The semifragile approach combines the properties of robust and fragile watermarks. Semifragile watermarks tolerate some degree of change (quantization noise from lossy compres- sion) to the watermarked digital content. The semifragile watermark can localize regions of dig- ital content that have been tampered, and it dis- tinguishes them from regions that are still authentic. Thus, a semifragile watermark can dis- tinguish between localized tampering and infor- mation-preserving, lossy transformations. The challenge of the evolution of watermarking is related to the information-preserving transfor- mations. Watermarks and attacks on watermarks are two sides of the same coin. A watermark’s goal is to be secured and robust enough to preserve the digital data’s value. However, watermark protec- tion’s goal is to be robust enough to resist attack but not at the expense of altering the value of the data being protected. On the other hand, the goal of the attack is to remove the watermark without destroying the protected data’s value. Scanning the issue Based on the experiences at four workshops on multimedia and security at the ACM Multimedia conference, the objective of this issue is to give an overview of current developments and problems in the field of multimedia and security. This spe- cial issue brings interesting and innovative papers from the ACM Multimedia 1999 Workshop on Multimedia and Security to a wider audience. The special issue analyzes specific security problems of multimedia systems and material in the digital environment. Based on our discussions in the workshops, we want to continue with state-of-the-art evaluation and discuss future needs for the design of multi- media security. Especially in the watermarking field, we need to evaluate the progress of robust- ness and of practical usage. With this special issue, we introduce the problems and general solution of multimedia security based on cryptography and digital watermarking. Out of various approaches, this issue is dedicated to improvements in digital watermarking through diversity and channel esti- mation. In the field of confidentiality, Jessica Fridrich et al. discuss a novel approach related to reliable detection of LSB steganography in gray- scale and color images. Tirkel and Hal raise the question of unique watermarks for every image. 20 1070-986X/01/$10.00 © 2001 IEEE Multimedia and Security Frederic Andres National Institute of Informatics, Japan Guest Editor’s Introduction