基于SVM和BP神经网络的车牌识别系统

针对车牌识别系统的车牌精确定位和车牌字符准确识别问题。提出一种基于SVM(支持向量机)和BP神经网络的车牌定位与识别算法。通过将HSV颜色空间和形态学方法相结合确定候选轮廓,以判断轮廓外接矩形的面积和长宽比筛选符合车牌特征的区域,并利用训练好的SVM模型对候选车牌区域进行测试判断,最终精确定位车牌的位置。此外,还可使用了BP神经网络进行车牌字符识别。经验证,该系统适用于复杂的车牌定位环境,且识别速度快,准确率高。     

基于图像处理的石英晶片定位技术研究

为了实现自动测量石英晶片的频率等相关参数并进行分档,研究了石英晶片的图像定位技术,主要利用图像处理技术对采集的图像进行图像二值化、边缘检测、轮廓提取等处理,完成对石英晶片中心的定位。经验证,该石英晶片定位技术能够在保证系统要求的石英晶片分选速度的前提下实现石英晶片的准确定位。     

一种基于LTE-MR劣化场景模式识别的自动优化方法

要进行LTE网络的精细优化必须依赖于对覆盖和干扰的地理分布精确掌握。本文基于MR数据建立精确定位干扰矩阵,形成从覆盖、干扰、用户、空间、时间等多维度定位网络问题的MR分析方法。 基于模式识别理论和MR数据的精确定位分析,智能识别出覆盖和干扰的场景,针对这些问题场景通过迭代算法自动生成相应的优化策略。     

基于机器视觉的工件定位和识别

根据机器视觉系统具有非接触性、实时性、稳定性及高精度的特点。文中采用机器视觉技术对工件的定位和识别进行了研究。通过求取工件轮廓中心矩和轮廓轴线二阶导数的最小值,求得工件的中心坐标和旋转角度;通过提取模板工件的Hu不变矩与工件轮廓的Hu不变矩,利用轮廓的相似度量函数识别工件类型。实验结果表明,通过提取工件轮廓矩能在旋转的情况下精确地定位和识别工件。     

基于OpenCV的石英晶片污垢检测技术研究

污垢检测是石英晶片缺陷检测的重要组成部分。为实现比较理想的污垢检测,采用基于开源计算机视觉库OpenCV的图像处理技术对石英晶片污垢缺陷进行检测。在此通过依次对图像进行平滑去噪、二值化阈值分割、轮廓提取和跟踪等处理,计算出轮廓的周长,将有缺陷与无缺陷晶片轮廓进行比较,为晶片污垢检测提供依据。实验结果表明在VisualStudio2008环境下,利用OpenCV库函数缩短了大量编程时间,提高了工作效率。     

一种新型晶片自动定位方法

阐述一种新型的硅片自动定位方法,并介绍了其特点和具体的实现方法。此方法可适应不同规格晶片的定位要求,实用性强。     

面向药瓶标签的字符定位方法研究

针对药瓶标签的特点,提出了一种数学形态学和投影特征相结合的药瓶标签字符定位方法。该方法首先对药瓶标签图像进行预处理,减小圆柱面造成的字符畸变程度,对处理得到的图像进行二值化和边缘检测等。其次对图像进行多次数学形态学处理,使图像分割成几个连通区域,根据先验知识以及轮廓跟踪方法得到字符所在的大致区域。最后利用水平投影和图像分割方法将粗定位得到的图像分割成三个单行字符,完成对字符区域的精确定位,便于后续的单个字符分割和识别。实验结果表明,该方法能对药瓶标签进行有效的定位。     

基于机器视觉的物体包装盒实时贴标系统设计

为了实现打印贴标机智能贴标的需求,提出了一种基于机器视觉的物体包装盒实时贴标系统设计方案,并完成了系统的软件设计和整套硬件测试。该系统的硬件部分主要用来获取图像数据以及进行打印贴标操作,软件部分采用基于轮廓提取和形状特征识别相结合的图像处理技术,能够完成对其输出图像进行目标识别和实时定位,并通过串口通信将数据转发给控制系统。实际应用表明,该系统具有操作简便、测试准确的特点,达到了设计要求。     

倒装热压键合设备中热气流对视觉定位系统影响研究

利用加热装置对晶片样片进行加热,采用机器视觉定位系统对晶片进行拍照采样,并利用图像处理软件进行位置识别,分析了在加热对图像定位系统的影响,并采取了措施降低热气流的影响,取得了比较满意的结果.     

圆环形压电振子疲劳仿真分析与实验研究

压电振子作为压电驱动机构的核心部件,其工作状态和疲劳强度直接决定系统的工作性能和使用寿命。该文利用ANSYS软件对圆环形压电振子进行疲劳仿真分析,得到振动模态云图和最大受力点。研究不同结构尺寸参数对压电振子疲劳寿命影响,并进行双晶片与单晶片压电振子相关数据对比。结果表明,金属材料65Mn的抗疲劳性能最好,减小振子直径或增加振子厚度能较好提高其疲劳寿命,而施加电压幅值过高会降低振子的使用寿命,双晶片压电振子的疲劳寿命比单晶片压电振子疲劳寿命长。对圆环形压电振子进行了疲劳寿命实验,研究表明,选取的此批振子完全能达到使用要求。     

Accurate estimation of statistical parametric variation in microelectronic test structures

Advanced technologies of microelectronic device fabrication need accurate determination of process parameters and their analysis. Test structures which represent individual process steps are incorporated in the design. Independent measurements can be carried out on these structures and the information is used as a feedback in order to achieve perfection at different stages of fabrication. This note presents an algorithm for analysis of measurement data and identification of defective sites on the wafer leading to an accurate yield analysis.     

基于坡度检测与点密度自约束调整的LOLA地理定位误差校正

月球轨道器激光高度计(Lunar Orbiter Laser Altimeter, LOLA)是目前月球探测中获取数据量最大、精度最高的激光高度计,可为其他月球产品提供地理控制。然而由于定轨不确定性和激光指向偏差,部分激光剖面存在地理定位误差。针对这些异常轨道数据,提出了一种改进的激光自约束调整方法。首先基于坡度将与周围地形差异过大的激光剖面识别为异常轨,并使用其余轨道作为参考对其进行基于点密度的自约束轨道调整(根据参考点密度自动选取合适的调整策略),最后合并两类轨道数据并进行最终调整,再将其作为改正后的结果。交叉点分析结果表明,该方法能有效校正异常轨道,从而提升数据精度。     

Hermetic-equivalent packaging of GPS MCM-L modules for highreliability avionics applications

Results are presented of comparative reliability testing of multichip modules (MCM's) fabricated with laminate substrates, and protected with various bare-die coatings. The demonstration MCM's included two design versions (flip-chip and wire-bond) of the digital portion of global positioning system (GPS) receiver multichip modules. This paper summarizes the results for the wire-bonded constructions. Standard encapsulants and new inorganic coatings (Dow Coming's ChipSeal(R) hermetic coating materials') were evaluated in environmental stress exposures corresponding to high reliability avionics applications. Full wafer probe testing was performed both before and after the supplemental ChipSeal processing and dip-chip wafer bump processing steps. ChipSeal and flip-chip wafer processing steps were shown to cause no yield degradation on wafer lots of five different IC types used in the overall program. The environmental test results demonstrate that MCM-L units with bare die packaging can be designed for very robust reliability applications such as military and other high reliability avionics     

A Feature-Scale Greenwood–Williamson Model for Metal Chemical Mechanical Planarization

In this work, a new feature-scale model is proposed for investigating the interaction between the wafer pattern and individual pad asperities in the process of chemical mechanical planarization (CMP). Based on the contact mechanics equation and the modified Greenwood–Williamson (GW) model which captures the evolution of feature curvature and the modification of the pad asperity height distribution, the discrete convolution and fast Fourier transform (DC-FFT) technique is adopted and combined with the Picard iteration method to calculate the direct contact pressure distribution between the wafer surface and the polishing pad. The computed pressure is then used to determine the local removal rate of the underlying patterns and predict the evolution of the wafer surface profile. Furthermore, the method is extended to capture the metal dishing as the feature size changes. It is shown that the present model can avoid the false simulated results produced by directly applying the original GW model for CMP when the feature size approaches zero. Otherwise, the calculated surface profile and dishing values of pattern geometries are in good agreement with the experimental data. Therefore, this model can not only be used to simulate the evolution of the wafer surface for global planarization at lower technology nodes, but can also be applied to provide some basic design rules for improving the process parameters and reducing the time and cost for developing new architectures.     

Electron-beam microcolumn fabrication and testing

In this paper, we report the improvement of a process for full microfabrication of miniaturized electron lenses specially design for low-energy (100 eV) lithography tools. The main advantages of this technique are the following. It is batch processing oriented, meaning that lenses can be easily built in a full wafer fabrication. With this procedure it is possible to develop a completely integrated process for machining arrays of lenses. Lens bores are aligned using an electron lithography process, resulting in highly accurate positioning. Finally, the source lens chip has not only one but several sets of lenses with different aperture-sizes, each producing a different beam diameter. A scheme is proposed with which the appropriate lens can be selected by means of a deflection system.     

A neural-network approach to recognize defect spatial pattern insemiconductor fabrication

Yield enhancement in semiconductor fabrication is important. Even though IC yield loss may be attributed to many problems, the existence of defects on the wafer is one of the main causes. When the defects on the wafer form spatial patterns, it is usually a clue for the identification of equipment problems or process variations. This research intends to develop an intelligent system, which will recognize defect spatial patterns to aid in the diagnosis of failure causes. The neural-network architecture named adaptive resonance theory network 1 (ART1) was adopted for this purpose. Actual data obtained from a semiconductor manufacturing company in Taiwan were used in experiments with the proposed system. Comparison between ART1 and another unsupervised neural network, self-organizing map (SOM), was also conducted. The results show that ART1 architecture can recognize the similar defect spatial patterns more easily and correctly     

采用晶圆传送机器人的晶圆预对准方法

针对传统的晶圆预对准控制系统成本高和体积大的不足，设计了基于晶圆传送机器人的晶圆预对准装置，并提出了高效、高精度的晶圆圆心和缺口定位算法。采用交换吸附的方式通过预对准装置一维旋转和晶圆传送机器人空间平移实现晶圆预对准。误差分析及预对准实验研究结果表明，晶圆圆心的定位精度＜50 ，预对准时间为10 s，满足设计要求。     

A model for rapid thermal processing: achieving uniformity throughlamp control

A first-principles approach to the modeling of a rapid thermal processing (RTP) system to obtain temperature uniformity is described. RTP systems are single wafer and typically have a bank of heating lamps which can be individually controlled. Temperature uniformity across a wafer is difficult to obtain in RTP systems. A temperature gradient exists outward from the center of the wafer due to cooling for a uniform heat flux density on the surface of the wafer from the lamps. Experiments have shown that the nonuniform temperature of a wafer in an RTP system can be counteracted by adjusting the relative power of the individual lamps, which alters the heat flux density at the wafer. The model is composed of two components. The first predicts a wafer's temperature profile given the individual lamp powers. The second determines the relative lamp power necessary to achieve uniform temperature everywhere but at the outermost edge of the wafer (cooling at the edge is always present). The model has been verified experimentally by rapid thermal chemical vapor deposition of polycrystalline silicon with a prototype LEISK RTP system. The wafer temperature profile is inferred from the poly-Si thickness. Results showed a temperature uniformity of ±1%, an average absolute temperature variation of 5.5°C, and a worst-case absolute temperature variation of 6.5°C for several wafers processed at different temperatures     

晶圆清洗过程中静电电压超标原因与改进

分析划片后的晶圆在清洗过程中静电电压严重超标的原因,提出4种改进方案,采用16种晶圆进行了验证。结果表明:清洗过程中关闭清洗机剥离胶膜用反冲气,操作人员佩戴防静电腕带,在取出晶圆过程中采用手持式可移动离子风枪对晶圆进行风淋,可解决长期存在的晶圆划片后清洗过程中表面静电电压严重超标的问题。     

Multivariate statistical methods for modeling and analysis of wafer probe test data

Probe testing following wafer fabrication can produce extremely large amounts of data, which is often used to inspect a final product to determine if the product meets specifications. This data can be further utilized in studying the effects of the wafer fabrication process on the quality or yield of the wafers. Relationships among the parameters may provide valuable process information that can improve future production. This paper compares many methods of using the probe test data to determine the cause of low yield wafers. The methods discussed include two classes of traditional multivariate statistical methods, clustering and principal component methods and regression-based methods. These traditional methods are compared to a classification and regression tree (CART) method. The results for each method are presented. CART adequately fits the data and provides a "recipe" for avoiding low yield wafers and because CART is distribution-free there are no assumptions about the distributional properties of the data. CART is strongly recommended for analyzing wafer probe data.