作者机构：
[李鹏; 王建新] School of Information Science and Engineering, Central South University, Changsha, 410083, China;[李鹏] School of Management and Information Engineering, Hunan University of Chinese Medicine, Changsha, 410208, China
通讯机构：
School of Information Science and Engineering, Central South University, Changsha, China
摘要：
A robust image hashing method based on radon transform and invariant features is proposed for image authentication, image retrieval, and image detection. Specifically, an input image is firstly converted into a counterpart with a normalized size. Then the invariant centroid algorithm is applied to obtain the invariant feature point and the surrounding circular area, and the radon transform is employed to acquire the mapping coefficient matrix of the area. Finally, the hashing sequence is generated by combining the feature vectors and the invariant moments calculated from the coefficient matrix. Experimental results show that this method not only can resist the normal image processing operations, but also some geometric distortions. Comparisons of receiver operating characteristic (ROC) curve indicate that the proposed method outperforms some existing methods in classffication between perceptual robustness and discrimination.
作者机构：
[丁长松; 王志英] College of Computer, National University of Defense Technology, Changsha, China;[胡志刚] School of Software, Central South University, Changsha, China;[丁长松] School of Management and Information Engineering, Hunan University of Chinese Medicine, Changsha, China
通讯机构：
College of Computer, National University of Defense Technology, Changsha, China
摘要：
A robust zero-watermarking algorithm is proposed based on merging features of sentences for Chinese text document authentication. In the scheme, a text is first segmented into sets of sentences, where a semantic code for every word can be obtained. Then the sentence entropy is calculated by the frequency of semantic codes, and the sentence relevance is calculated by the semantic similarity between words through the tree structure of words in Tongyici Cilin. By employing the sentence entropy, the sentence relevance, and the sentence length, a weighting function is used to obtain the final weight of each sentence. The nouns and verbs of the high weight sentences are selected to construct a watermark, which is encrypted and registered with a trusted third party called Certificate Authority (CA). To resolve disputes, the similarity between the watermark generated from the suspicious text and the watermark from CA is calculated. The experimental results show that the proposed algorithm offers better performance in terms of imperceptibility and robustness than other available algorithms.
摘要：
Communications via instant message tools have become increasingly popular in people's daily lives. However, one of the main issues in communication is the transmission of secret information. There are many methods for covert communications. In this paper, a novel text steganography method in chat is proposed which utilizes emoticons and interjections. Due to the tremendous numbers of emoticons and interjections used in many chat tools, the pre-shared sets of emoticons and interjections can be enlarged as required. Then, through selecting the emoticons and interjections of corresponding encoding in different locations, secret information can be embedded into the chat text. Experiment results demonstrate that the capacity and embedding efficiency have better performance than other text steganography methods in chat.
关键词：
reversible data hiding;medical images;ROI-based;prediction error expansion;sorting
摘要：
A novel ROI-based reversible data hiding scheme is proposed for medical images, which is able to hide electronic patient record (EPR) and protect the region of interest (ROI) with tamper localization and recovery. The proposed scheme combines prediction error expansion with the sorting technique for embedding EPR into ROI, and the recovery information is embedded into the region of non-interest (RONI) using histogram shifting (HS) method which hardly leads to the overflow and underflow problems. The experimental results show that the proposed scheme not only can embed a large amount of information with low distortion, but also can localize and recover the tampered area inside ROI.
期刊：
Journal of Computational and Theoretical Nanoscience,2015年12(10):3658-3661 ISSN：1546-1955
通讯作者：
Peng, Yingying
作者机构：
[Peng, Yingying; Li, Man] College of Management and Information Engineering, Hunan University of Chinese Medicine, Changsha, Hunan, China;[Li, Kenli; Peng, Yingying] College of Information Science and Engineering, Hunan University, Changsha, Hunan, China
通讯机构：
College of Management and Information Engineering, Hunan University of Chinese Medicine, Changsha, Hunan, China
关键词：
Cluster;Data Mining;Improved K-Means
摘要：
K-Means algorithm has been researched adequately in recent years. Clustering result of traditional K-Means algorithm is affected by the choice of initial point and noise. In addition to, traditional K-Means algorithm only favors clusters with spherical shapes and similar sizes. A novel K-Means algorithm combining K-Means algorithm and KNN algorithm called KK-Means is proposed to solve these weaknesses in this paper. Experimental result shows that KK-Means algorithm has better performance more than traditional K-Means algorithm.
作者机构：
[石继连; 田静; 王海琴] College of Pharmacy, Hunan University of Chinese Medicine, Changsha 410208, China;Hunan Province 12-5 Provincial Key Disciplines of Pharmacy, Changsha 410208, China;[吴春英] College of Management and Information Engineering, Hunan University of Chinese Medicine, Changsha 410208, China;[杨岩涛; 刘文龙; 贺福元] College of Pharmacy, Hunan University of Chinese Medicine, Changsha 410208, China, Hunan Province 12-5 Provincial Key Disciplines of Pharmacy, Changsha 410208, China
通讯机构：
[He, F.-Y.] C;College of Pharmacy, Hunan University of Chinese Medicine, Changsha 410208, China
作者机构：
[唐宇; 吴德智] College of Pharmacy, Hunan University of Chinese Medicine, Changsha, 410208, China;Key Laboratory of Property and Pharmacodaynamics for TCM, State Administration of Traditional Chinese Medicine, Changsha, 410208, China;Pharmaceutical Preparation Technology and Evaluation Laboratory of TCM, Hunan University of Tradition Chinese Medicine, Changsha, 410208, China;[吴春英] College of Management and Information Engineering, Hunan University of Traditional Chinese Medicine, Changsha, 410208, China;Supermolecular Mechanism and Mathematic-Physics Characterization for Chinese Materia Medica, Hunan University of Traditional Chinese Medicine, Changsha, 410208, China
通讯机构：
[He, F.-Y.] C;College of Pharmacy, Hunan University of Chinese Medicine, Changsha, China
关键词：
Single Nucleotide Polymorphism;tagSNPs;Genetic Algorithm;Artificial Neural Network
摘要：
Currently, many approaches have been developed to be applied in the tagSNP selection research. However, there are still drawbacks existing in these methods, manifested chiefly by high time complexity, large number of selected tagSNPs, low prediction accuracy and inefficient tagSNPs in followup study. We propose an informative SNP selection method framework based on genetic algorithm in this paper to address these problems. In this study, we separately improve the phases of informative SNPs set construction and haplotypes reconstruction. Firstly, we eliminate the large number of redundant SNPs with LD values to obtain a candidate subset with small redundancy, and then seek for optimization with genetic algorithm not only to ensure the reconstruction accuracy efficiently but also to reduce time complexity greatly. Besides, to avoid retrain prediction model repeatedly by traditional methods (e.g., MLR, SVM and so on), we make full use of BP neural network's multiple output characteristics to reconstruct all non-tagSNP at once. Thus, it significantly saves computational complexity. The experimental results show that our method performs much better than the current dominating tagSNP selection methods.
期刊：
Computer Modelling and New Technologies,2014年18(12):245-249 ISSN：1407-5806
通讯作者：
Peng, Yingying
作者机构：
[Ren, Xuegang; Peng, Yingying] Department of Management and Information Engineering, Hunan University of Chinese Medicine, Changsha, Hunan, China;[Hu, Defa] School of Computer and Information Engineering, Hunan University of Commerce, Changsha, Hunan, China
摘要：
Aiming at the Low-Density Parity-Check Codes, a reliability-based multibit-flipping decoding algorithm is proposed in the paper. The multibit-flipping criterion is based on the reliable bit position and the threshold in the flipping-decision (number of flipping bits) can be dynamically adjusted during the decoding process. The proposed algorithm is on the basis of the belief propagation decoding algorithm, and then can be derived from its theory. Compared with the traditional weighted bit-flipping decoder and the multi-bit flipping decoder, the proposed decoder can provide a faster converges faster convergent rate and better performances. Simulation results demonstrate that the proposed algorithm achieves a better balance between performance and complexity.
作者机构：
[丁长松] School of Administration and Information Engineering, Hunan University of Chinese Medicine, Changsha 410208, China;[胡志刚] School of Software, Central South University, Changsha 410083, China;[丁长松; 王志英] College of Computer, National University of Defense Technology, Changsha 410073, China
通讯机构：
School of Administration and Information Engineering, Hunan University of Chinese Medicine, China
