基于鲁棒特征挖掘的图像副本检索

为了提高视觉特征的鲁棒性,提出尺度空间下最稳定极值区域特征的挖掘算法.该算法在尺度空间下检测特征,首先通过分析特征的局部稳定性来挖掘鲁棒的特征,然后对区分力低的特征基于熵值进行过滤;在此基础上,基于尺度空间下最稳定极值区域特征的尺度一致性提出了基于最大匹配尺度的检索算法,可提高局部敏感哈希高维索引的检索性能.与多种已有方法进行比较的结果表明,文中算法的平均检索精度相对提高10％以上,查询效率也有提升.     

Curvature estimation for meshes based on vertex normal triangles

The estimation of surface curvature is essential for a variety of applications in computer graphics because of its invariance with respect to rigid transformations. In this article, we describe a curvature estimation method for meshes by converting each planar triangular facet into a curved patch using the vertex positions and the normals of three vertices of each triangle. Our method interpolates three end points and the corresponding normal vectors of each triangle to construct a curved patch. Then, we compute the per triangle curvature of the neighboring triangles of a mesh point of interest. Similar to estimating per vertex normal from the adjacent per triangle normal, we compute the per vertex curvature by taking a weighted average of per triangle curvature. Through some examples, we demonstrate that our method is efficient and its accuracy is comparable to that of the existing methods.     

On constant-round concurrent non-malleable proof systems

Security under man-in-the-middle attacks is extremely important when protocols are executed on asynchronous networks, as the Internet. Focusing on interactive proof systems, one would like also to achieve unconditional soundness, so that proving a false statement is not possible even for a computationally unbounded adversarial prover. Motivated by such requirements, in this paper we address the problem of designing constant-round protocols in the plain model that enjoy simultaneously non-malleability (i.e., security against man-in-the-middle attacks) and unconditional soundness (i.e., they are proof systems).We first give a construction of a constant-round one-many (i.e., one honest prover, many honest verifiers) concurrent non-malleable zero-knowledge proof (in contrast to argument) system for every NP language in the plain model. We then give a construction of a constant-round concurrent non-malleable witness-indistinguishable proof system for every NP language. Compared with previous results, our constructions are the first constant-round proof systems that in the plain model guarantee simultaneously security against some non-trivial concurrent man-in-the-middle attacks and against unbounded malicious provers.     

An extended matrix encoding algorithm for steganography of high embedding efficiency

F5 steganography is much different from most of LSB replacement or matching steganographic schemes, because matrix encoding is employed to increase embedding efficiency while reducing the number of necessary changes. By using this scheme, the hidden message inserted into carrier media imperceptibly can be transferred via a more secure subliminal channel. The embedding domain is the quantized DCT coefficients of JPEG image, which makes the scheme be immune to visual attack and statistical attack from the steganalyst. Based on this effective scheme, an extended matrix encoding algorithm is proposed to improve the performance further in this paper. By changing the hash function in matrix encoding and converting the coding mode, the embedding efficiency and embedding rate get increased to a large extent. Eventually, the experimental results demonstrate the extended algorithm is superior to the classic F5 steganography.     

Fuzzy identity based signature with applications to biometric authentication

We introduce a new cryptographic primitive which is the signature analog of fuzzy identity based encryption (FIBE). We call it fuzzy identity based signature (FIBS). It possesses similar error-tolerance property as FIBE that allows a user with the private key for identity ω to decrypt a ciphertext encrypted for identity ω′ if and only if ω and ω′ are within a certain distance judged by some metric. We give the definition and security model of FIBS and present the first practical FIBS scheme. We prove that our scheme is existentially unforgeable against adaptively chosen message attack in the standard model. To our best knowledge, this primitive was never considered in the identity based signature before. FIBS is of particular value for biometric authentication, where biometric identifiers such as fingerprints, iris, voice and gait are used in human identification. We demonstrate the applicability of our construction to secure biometric authentication.     

基于GL847的扫描图像校正

扫描仪在扫描成像过程中,诸多因素会给扫描图像带来误差和扭曲,因而必须对其加以校正。在分析扫描仪误差产生原理的基础上,详细讨论了基于GL847的扫描图像增益与偏移校正机制,提出了一种有效的扫描图像校正参数获取方法。     

一种视频序列拼接的新方法

提出了一种基于视频序列拼接的新方法。首先,利用KLT算法对视频序列中特征点进行提取和跟踪,实现关键帧粗略选取;其次,在选取的关键帧中利用SURF算法进行特征提取,利用最近邻距离比进行特征点匹配,通过RANSAC估计算法求精单映矩阵,并结合关键帧选取判定寻找最佳关键帧;最后,利用级联单映矩阵和加权融合算法实现视频序列拼接。实验验证了该方法的有效性。     

基于MSP430F149与Si4432的无线传感器网络的实现方法

介绍了基于MSP430F149与Si4432的无线传感器网络节点的硬件及软件设计方法,给出了主要的硬件电路和软件流程图,提出了一种简单易实现的低功耗自组织协议。实验证明,该无线传感器网络具有组网速度快、功耗低、抗干扰能力强、通信稳定可靠等特点,可应用于多个领域的信息探测,具有广阔的市场前景。     

一种新型的毫米波紧凑波导双工器

利用电路等效分析法结合全波建模提出具有宽带工作特性的改进型H面T型结。结合交叉耦合原理和一腔多模理论,通过S参数的多项式综合和矩阵消零的方法提取滤波器耦合矩阵。设计了新型结构的双模交叉耦合毫米波双工器。测试结果表明,相比于传统的单模直接耦合双工器,这种双工器不但尺寸更加紧凑,而且具有更好的频率选择性及隔离度等优点。     

石脑油裂解过程中一次反应选择性系数的调整方法研究

在分析石脑油裂解炉辐射段的Kumar分子反应动力学模型的基础上,提出了一种应用数据融合技术对Kumar模型一次反应选择性系数进行在线调整的方法。首先,利用进化的遗传算法,以改进的石脑油裂解模型计算出的产率与实际工况产率的误差作为目标函数,对选择性系数进行离线调整,进而建立原料性质与选择性系数相匹配的标准样本数据库。其次,利用数据融合技术中的Dempeter-Shafer证据理论建立模型,计算待估物料与标准样本数据库中样本物料的匹配度,进而对待估原料的一次反应选择性系数进行估算。研究结果表明,利用估算出的一次反应选择性系数建立的裂解工艺模型具有较高精度,将仿真结果与实际工况产率进行对比,证实了该方法的有效性和可行性。