Color texture segmentation based on image pixel classification

Image segmentation partitions an image into nonoverlapping regions, which ideally should be meaningful for a certain purpose. Thus, image segmentation plays an important role in many multimedia applications. In recent years, many image segmentation algorithms have been developed, but they are often very complex and some undesired results occur frequently. By combination of Fuzzy Support Vector Machine (FSVM) and Fuzzy C-Means (FCM), a color texture segmentation based on image pixel classification is proposed in this paper. Specifically, we first extract the pixel-level color feature and texture feature of the image via the local spatial similarity measure model and localized Fourier transform, which is used as input of FSVM model (classifier). We then train the FSVM model (classifier) by using FCM with the extracted pixel-level features. Color image segmentation can be then performed through the trained FSVM model (classifier). Compared with three other segmentation algorithms, the results show that the proposed algorithm is more effective in color image segmentation.     

Q345B钢动态再结晶动力学模型研究

在Gleeble1500热模拟机上进行Q345B钢单道次压缩变形实验,得到其真应力-真应变曲线,结合加工硬化率曲线,确定了Q345B钢动态再结晶临界应变εc、峰值应变εP和稳态应变εs。根据实验结果得到Zener-Hollomon方程和动态再结晶状态图,利用Johnson-Mehl-Avrami（JMA）方程法得到再结晶体积分数实际值,采用3种不同的再结晶体积分数预报模型对实验数据进行回归,并对再结晶体积分数实测值和预报值进行对比。结果表明,Epsilon-S／Epsilon-C模型精度最高,Epsilon-S模型精度次之,Epsilon-P模型精度最差。     

结合SURF描述符和广义近邻图的图像配准算法

针对传统互信息配准方法计算量较大且未利用图像空间信息的缺点,提出了一种结合SURF描述符和广义近邻图的图像配准算法。该算法用SURF从图像中提取尺度空间特征点并获得特征点描述子,然后用广义近邻图来估计Rényi熵与互信息。该算法结合了SURF描述子的鲁棒性和广义近邻图估计Rényi熵的高效性。实验结果表明,对于真实遥感图像,该算法在配准准确度、鲁棒性和速度上都明显优于几种传统配准方法。     

虚拟修模技术在涡轮叶片模具优化设计中的应用研究

为了提高涡轮叶片的成形质量,减少模具的修模次数,提出了一种基于反变形原理的虚拟修模方法。通过使用P ROCAST软件对涡轮叶片铸造过程进行仿真,利用有限元模型计算涡轮叶片的精铸位移场,建立基于位移场的虚拟修模方法,并使用该方法对模具进行优化设计,经涡轮叶片模具设计仿真实验验证,所采用的方法取得了良好的效果,可初步应用于模具的优化设计。     

A CAM-based keyword match processor architecture

This paper demonstrates a keyword match processor capable of performing fast dictionary search with approximate match capability. Using a content addressable memory with processor element cells, the processor can process arbitrary sized keywords and match input text streams in a single clock cycle. We present an architecture that allows priority detection of multiple keyword matches on single input strings. The processor is capable of determining approximate match and providing distance information as well. A 64-word design has been developed using 19,000 transistors and it could be expanded to larger sizes easily. Using a modest 0.5 μm process, we are achieving cycle times of 10 ns and the design will scale to smaller feature sizes.     

基于融合分析的感应电机轴承故障检测方法

根据感应电机轴承发生故障时的振动信号特性以及定子电流特性，求出三相电流的Park矢量模信号，并将其与电机滚动轴承振动信号经解调处理后的包络信号进行融合分析。可以从振动信号与电流信号的融合谱图中有效地提取轴承故障特征信息，并将其作为故障识别的依据。实验结果表明，本文检测方法具有较高的信噪比，提高了诊断的可靠性。     

企业应用环境系统升级方案研究

本文以一家企业实施系统升级项目为例,从系统域控制及邮件系统集成升级、防火墙设立提供安全访问控制、防病毒及垃圾邮件系统、备份、软件分发管理等全方位探讨了企业如何通过系统整合,实现以管理软件系统为核心的企业应用。     

TiO2/ZrO2复合陶瓷膜的性能研究

采用反相微乳液法制备TiO2/ZrO2复合微乳液,并以其作为前驱体制得TiO2/ZrO2复合膜。研究了复合膜的机械性能、表面形貌及杀菌性能。由扫描电子显微镜观察到TiO2/ZrO2的复合膜中微粒以棒状形式分布。杀菌实验测试结果表明其对大肠杆菌有很好的杀菌效果,灭菌率达98%~99%。     

由状态变量识别板梁桥上移动荷载

将板梁桥等效为正交异性板,车辆等效为等间距常速度在正交异性板上移动的一组荷载。根据弹性板理论、HALMITON'S原理和模态叠加原理,得到了系统状态方程,并基于规则化技巧推导了状态空间移动荷载识别方法。数值模拟和试验结果表明,该方法用挠度或应变识别车辆的轴重和轮载都是有效、可行的。     

Influence of TiO<Subscript>2</Subscript> on the alkali resistance and structure of invert glass

Through measuring the alkali resistance of the invert glass and the quanity of SiO₂ and TiO₂ migrating fron the glass into the solution, the influence of TiO₂ on the alkali resistance of the glass is discussed and its structure is also analyzed by infrared spectroscopy. It is concluded that TiO₂ has double functions for the alkali resistance of the invert glass. On the one hand, both TiO₂ polarizing the secondary ions in glass and TiO₂ isomorphism replacement of SiO₂ make the alkali resitance of the glass decrease. On the ther hand, TiO₂ patching network and anti-erosion covering help to increase the alkali resistance. ZHAO Qing-lin: Born in 1972 Funded by the National Natural Science Foundation of China(No. 59372081)