文献类型：会议论文

英文题名：Robust image watermarking scheme with general regression neural network and FCM algorithm

作者：Jing, Li[1,2];Liu, Fenlin[1];Liu, Bin[1]

第一作者：Jing, Li;景丽

通讯作者：Jing, L[1]

机构：[1]ZhengZhou Informat Sci & Technol Inst, Zhengzhou 450002, Peoples R China;[2]Henan Univ Finance & Econ, Coll Informat, Zhengzhou, Peoples R China

第一机构：ZhengZhou Informat Sci & Technol Inst, Zhengzhou 450002, Peoples R China

通讯机构：[1]corresponding author), ZhengZhou Informat Sci & Technol Inst, Zhengzhou 450002, Peoples R China.

会议论文集：4th International Conference on Intelligent Computing

会议日期：SEP 15-18, 2008

会议地点：Shanghai, PEOPLES R CHINA

语种：英文

外文关键词：Computation theory - Fuzzy inference - Image compression - Image processing - Image watermarking - Intelligent computing - Neural networks - Regression analysis - Watermarking

摘要：A robust image watermarking method based on general regression neural network (GRNN) and fuzzy c-mean clustering algorithm (FCM) is proposed. In order to keep the balance between robustness and imperceptible, it uses FCM to adaptively identify watermarking embedding locations and strength based on four characteristic parameters of human visual system. For good learning ability and fast train speed of GRNN, it trains a GRNN with the feature vector based on the local correlation of digital image. Then embed and extracted watermark signal with the help of the trained GRNN. In watermark extracting it does not need original image. Experimental results show that the proposed method has better performance than the similar method in countering common image process, such as Jpeg compression, noise adding, filtering and so on.