A neural network model for fault detection in conjunction with a programmable logic controller

This paper discusses the feasibility of using neural networks as a tool in the fault detection process. A neural network is integrated with a state language programmable logic controller, an important device in an automatic control system. Time series data related to time spent in a state is gathered and used as input into a neural network, for the purpose of identifying when a fault has occurred. A feedforward neural network is used to identify which (if any) of three types of faults may have occurred. Experimental results related to sensitivity and accuracy measures are presented. A brief review of related applications and research is also presented.     

Sub-optimal solutions to track detection problem using graph theoretic concepts

This paper demonstrates how the problem of tracking targets, which appear as either straight or curved lines in two-dimensional display images (or data images) can be formulated in terms of a directed weighted graph model and how dynamic programming techniques can be efficiently applied to reach an optimal or sub-optimal solution. In general, track detection algorithms providing optimal solutions have good detective ability, but most of them suffer from the inability to detect discontinuous lines or to resolve efficiently pairs of crossing lines. A sub-optimal solution is provided that efficiently overcomes these weaknesses. We focus on modeling the track detection problem in terms of a graph, formulating fast sequential/parallel sub-optimal track detection algorithms and testing them on simulated data in order to show their detective ability. Moreover, we specify the conditions under which sub-optimal algorithms can perform at least as well as their corresponding optimal algorithms. This is significant for the track detection problem where fast, accurate and real-time detection is considered a necessity.     

A protocol for performance evaluation of line detection algorithms

Accurate and efficient vectorization of line drawings is essential for any higher level processing in document analysis and recognition systems. In spite of the prevalence of vectorization and line detection methods, no standard for their performance evaluation protocol exists. We propose a protocol for evaluating both straight and circular line extraction to help compare, select, improve, and even design line detection algorithms to be incorporated into line drawing recognition and understanding systems. The protocol involves both positive and negative sets of indices, at pixel and vector levels. Time efficiency is also included in the protocol. The protocol may be extended to handle lines of any shape as well as other classes of graphic objects.     

螺杆泵的检测方法及检测标准制定

螺杆泵检测是避免不合格螺杆泵下井、螺杆泵早期失效和提高螺杆泵使用寿命与采油效益的重要技术手段。重点介绍了模拟不同工况的螺杆泵工作特性检测系统及检测定子通径为10－60mm的定子衬套偏心度检测仪的组成及其检测方法和原理。针对大庆油田的实际情况，在进行大量室内和现场试验的基础上，制定了GLB120－27、GLB40－42和GLB500－14型螺杆泵的检测标准。检测技术及其标准实施以来见到了明显效果。     

Edge detection,three-dimensional cell boundary reconstruction and volume and surface area estimation from differential interference contrast images

Three-dimensional (3-D) cell morphology is important for the understanding of cell function and can by quantified in terms of volume and surface area. Differential interference contrast (DIC, or Nomarski) imaging can enable cell edges to be clearly visualized in unstained tissue due to the slight difference in refractive index between aqueous media and cytoplasm. DIC is affected in only one direction - the direction of the optical shear. A 1-D edge detector was used in that direction with a scale length equal to that of an in-focus edge to highlight cell boundaries. By comparison with the signal from the edge detector on an out-of-focus slice, the in-focus slices could be segmented and, after noise suppression, cell outlines obtained. A voxel paradigm was used to calculate cell volume and differential geometry was used for surface area estimation. We applied this approach to obtain 3-D dimensional information by optical sectioning of motile Amoeba proteus.     

基于运动矢量的视频图象数字水印算法

数字水印技术是近年发展起来的一种用于数字产品版权保护和真伪性论证的新兴技术。与静止图象相比，视频水印技术要满足盲检的要求。本文根据运动矢量的特征值ρl，提出了一种视频图象水印新算法。实验结果表明，所提出的水印算法简单、快速，能满足视频编码的实时性要求；算法与现有的视频压缩标准有很好的兼容性；水印的提取具有盲检功能，无需原图象；水印的嵌入不影响I帧的图象质量，与嵌入水印前的原压缩图象相比，嵌入水印后的视频图象信噪比损失很小。     

标准化钻速法检测地层压力技术的改进与应用

利用钻井资料检测地层压力的方法，基本上都是以泥岩的压实规律为基础，以钻速的变化为根据的。但是在钻井过程中影响钻速的因素很多，可以衡量一种检测方法优劣，主要是看其能否合理、全面地考虑诸多影响因素，能否适用于现场。本文介绍标准化钻速法检测地层压力的基本原理和工作步骤，提出了适用于现场的改进的数学模型和应用方法，并与实测值很接近。     

利用梯度方向信息的随机Hough变换

针对随机Ｈｏｕｇｈ变换（ＲＨＴ）在处理复杂图像随机采样造成的大量无效积累提出一种改进的用于直线检测的ＲＨＴ，较发地解决了无效果累积问题，并使改进后的算法具有计算速度更快，占用内存更少以及检测性能更好等优点。     

基于空间矩的灰度边缘亚像元度量精度分析

灰度图像中边缘定位受采样密度限制，亚像元精度度量方法在需要高精度的应用中受到重视，要想知识到底需要多高的分辨率才能达到定精度指标，就有必要分析特征定位方法的最坏情况的精度，文中用分析和实验的方法给出了基于空间矩的边缘定位方法，对于边缘点，边缘直线以及直边缘区域的最坏情况定位精度分析。分析和讨论的结果不但可以用来确定已知形状位置估计的最大误差，而且还可用于对准的形状几何参数，以获得较高的精度。