978-1-4244-9983-0/11/$26.00 ©2012 IEEE 1 Multiple Watermarking On Digital Medical Images for Tamper Detection and Integrity Control M.T. Kurniawan Graduate School and Faculty of Industrial Engineering, Telkom Institute of Technology, Bandung, Indonesia E-mail: ujangtegoeh@gmail.com Adiwijaya Faculty of Science, Telkom Institute of Technology Bandung, Indonesia E-mail: adiwijaya@ittelkom.ac.id Wiseto Agung R&D Center, PT. Telekomunikasi Indonesia Bandung, Indonesia E-mail: wiseto.agung@gmail.com Abstract—In the current digital era, patient data in the form of digital medical images in several hospitals are widely used. There are two important thin to use of digital medical images namely the ownership authority (integrity control) and the authenticity of the image (authentication), because digital medical images are very easily manipulated. To maintain the authority of ownership, it needs robust watermarking techniques in which the embedded data is not easily damaged if the image has been manipulated. Meanwhile, to detect the authenticity of the image, it needs fragile watermarking technique in which the embedded data is easily damaged if the image has been manipulated. In this paper, we implement Reed-Solomon code for robust watermark in wavelet domain and SHA-256 for fragile watermark in Hash Block Chaining. The proposed multiple watermarks can be implemented simultaneously on an image so that the integrity control and authenticity of the image detection can be applied at once. Keywords-component; Multiple watermarking, medical images, robust watermarking, fragile watermarking I. INTRODUCTION Nowadays, patient’s data in the hospital can be stored in electronic media. The data in digital medical images form such as X-ray image, mammogram form and others can be very easily manipulated by the rapid development of information technology today. Medical images in digital form must be stored well to preserve stringent image quality standards and prevent unauthorized disclosure of patient data [4]. There are two important things must be concerned in digital medical images such as the authority of ownership (Integrity control) and the authenticity of the image (Authentication). As consequences to ward these cases it is necessary to apply watermarking techniques. The principle of watermarking is embedding digital data (either text or image) into the original digital medical image to meet the needs of integrity control (to maintain the authority of ownership) and Authentication (to detect the authenticity of the image). . Multiple watermarks has two parts: signature watermark in the form of robust watermark and reference watermark in the form of fragile watermark [4]. Signature watermark is used to maintain the authority of ownership because it has robust characteristic that is not prone to damage if the embedded image is manipulated so the data remains safe. The reference watermark is used to detect the authenticity of the image. This reference watermark is highly vulnerable to the manipulation of imagery, however due to its fragility that will easily detect the manipulation of the image so that the authenticity of the image can be maintained [1]. Methods used to embed a watermark in digital medical images are very diverse. The first is a watermark on the image method which does not allow embedding in the image areas that are considered important (Region of Interest). Although this method produces good image quality in the ROI area, but the main problem is that it is easy to do copy attack in the area (areas that are not embedded watermark). II. MULTIPLE WATERMARKING Multiple watermak has two watermark namely robust watermark for integrity control and fragile watermark for tamper detection. There are several research in mutliple watermarking. Woo, et al. [4] proposed method multiple watermarking on digital medical image which is suitable for privacy control and tamper detection in medical images. To provide data security and patient privacy, patient information embedded into an annotation watermark. This annotation watermark is embedded into RONI (region of non interest) image using a robust embedding method. Then, it is embedded using a linear additive method into the three high pass bands of discrete wavelet transform (DWT) of original image border or RONI. And to provide integrity of the medical image can be authenticated using a fragile watermark. This fragile watermark is embedded into the ROI (region of interest) image using the least significant bit (LSB) method. Kallel, et al. [1] proposed to use the following scheme in order to preserve the image history in the digital medical field. This method divided into two parts. The first one is to embed the patient’s diagnoses in the digital medical images