成捆原木计算机图像检尺系统研究及应用

主要论述基于成捆原木端面计算机图像处理技术的原木自动检尺系统的组成,研究并设计了系统主要图像处理和识别算法,实现了木捆中每根原木端面轮廓识别,并计算了检尺参数.     

等长堆积原木材积的自动检测

等长且堆放成捆的原木材积由人工检尺测得，会产生很大误差。本文提出原木材积测量的新方法，该方法采用计算机视觉技术，对摄像机摄取的原木捆端面图像进行处理，自动计算出截面积值，从而得出木材捆的材积值。     

基于计算机视觉的原木高出材率的研究

计算机能模拟人眼视觉环境来甄别、测量视野中的事物,随着精度的不断提高,计算机视觉能代替人眼实现简单、重复的人工操作。计算机视觉结合原木剖料可以提高原木的出材率,减少木材损耗。凭借计算机的高效性、准确性,计算机视觉能实现原木的最大利用,最大程度减少生成方木浪费的原料, 提高原木出材率。将所提算法应用于自动化带锯原木切割系统的基本流程是通过图像预处理来消除图像噪声,通过颜色分割去除背景,通过边缘检测给出感兴趣区域的轮廓,应用形态学运算连接和填补被误处理的轮廓边缘,计算最大面积的拟合椭圆。实验结果表明,该算法可以满足实际生产的需要,实验精度达到95%以上。     

基于圆形约束CV-LIF模型的原木端面图像分割

针对自然条件下原木端面图像的分割问题,结合原木端面图像的特点,改进传统CV（Chan and Vese）模型,对演化曲线内部使用梯度进行拟合,同时融入局部图像拟合LIF（Local Image Fitting）模型,加入圆形先验知识,提出了基于圆形约束的改进活动轮廓模型CV-LIF,将全局能量和局部能量结合到一起,共同约束轮廓线的演化。在对图像进行预分割的基础上,利用多水平集表示待分割区域,运用基于圆形约束的改进活动轮廓模型对每个水平集区域进行再分割,解决了复杂背景下多个原木端面分割不准确的问题。通过实验,分别对单个及多个原木端面图像进行分割,结果表明该方法可以较好地分割出图像中的原木端面,而且具有较好的抗噪性能,实现速度较快。     

一种色彩特征的自动提取算法及其软件实现

色彩特征的提取在图像识别与图像检索中都起着至关重要的作用.以往的特征提取主要是靠人工,或者是通过有监督的学习来实现,这就限制了整个识别或检索过程的自动化.针对BMP图像的特点,巧妙利用图像色彩空间之间的转换,提出了一种基于数据聚类的主色提取算法,并在Visual C 6.0环境下对其进行了软件实现.实验结果表明,该算法实现了图像预处理的基本操作,并通过适当的数据聚类算法,实现了BMP图像色彩特征的自动提取.同时该方法较传统方法的优越性以及软件的实用性,为整个系统的自动化以及下一步的研究奠定了基础.     

一种基于图像处理的成捆钢筋计数方法

本文研究了一种基于图像处理的成捆钢筋快速计数方法。首先对生产出的成捆钢筋端面进行摄像;然后对拍摄的端面图像进行去噪、二值化、数学形态学等处理,得到图像轮廓;再对图像轮廓进行提取分析,计算出各个轮廓区域中包含的钢筋数目,从而统计出钢筋的总根数。实验结果表明,该方法应用于实际生产中能降低成本,提高计数精度。     

彩色图像序列中运动人体轮廓提取

在视频序列的人体运动分析中,实时提取出运动人体轮廓,是很多研究起始的关键步骤.而彩色图像由于具有比灰度图像更多的视觉信息,受到了越来越多的重视.采用了一种新的色彩背景模型;运用改进的背景差分方法在复杂背景下获得运动人体的轮廓.实验结果表明上述算法对噪声抑制和人体图像断裂处填充都是有效的,在目标物是运动物体,且背景绝大多数均为静止时,该算法适用,能够实时提取出运动人体的轮廓.     

基于轮廓线与特征融合的电网变电设备三维自动化运检方法

因工作环境的复杂性,变电设备成为智能电网中故障频率最高的装置之一,易发生设备接地故障、保险熔断故障、绝缘材料老化等问题,为提高电网变电设备三维自动化运检精度,提出基于轮廓线与特征融合的电网变电设备三维自动化运检方法研究。采集电网变电设备三维图像,获取电网变电设备三维图像初始轮廓点,筛选变电设备轮廓点,依照顺序连接提取出的关键轮廓点,得到变电设备三维图像轮廓线,结合形态学滤波算法,对图像进行开、闭运算,依据轮廓线的闭合情况,分割出目标图像与背景图像,提取包括HOG特征与LBP特征的变电设备目标图像特征,经过特征融合后,构造最优分类超平面,制定电网变电设备三维自动化运检规则,判定变电设备运维情况,最终实现变电设备的三维自动化运检。实验数据显示：该方法识别出变电设备的正常与异常状态,在不同实验工况背景下,应用提出方法获得的变电设备运检精度达到了96%。提高了变电设备的自动化运检的识别及运检精度,满足现今变电设备的运检需求。     

基于局部分形维数特征的医学超声图像轮廓提取

医学超声图像轮廓提取的研究对疾病诊断有重要意义,传统方法受医学超声图像噪声影响较难获得准确的提取结果。本文研究通过引入局部分形维数作为图像轮廓特征,对医学超声图像进行轮廓提取。通过与传统方法比较,实验结果表明本文方法对医学超声图像中存在的噪声有较好的抗干扰能力,能够在噪声背景中较好提取图像的轮廓,为医学超声图像轮廓提取开辟了一条新的途径。     

基于凹点搜寻的棒材计数方法研究

针对现有的成捆棒材计数方法存在计算速度低、色差偏差大、计算量大、易受边缘模糊及残缺影响的问题,提出了一种通过对成捆棒材中的黏连棒材进行分割,然后统计连通区域,从而达到棒材计数目的的新方法。该方法首先对成捆棒材的图像进行预处理,利用Canny边缘检测获得黏连棒材的边缘轮廓,然后在此轮廓上扫描凹区域,将临近的凹点作为一个凹点群,取其中最突出的点作为此处凹点;再利用凹点匹配条件进行匹配,以匹配成功的凹点对作为界限来分割图像,最后统计分割后的连通区域,即达到计数的目的。实验结果表明,该方法对于黏连棒材的分割特别准确,分割后的图像边缘与成捆棒材的实际边缘很贴近,在无互相遮挡的情况下,直径为12 mm以上的钢筋的分割准确率达到99.99%以上,直径为10~12 mm的钢筋的分割准确率在99.80%以上,达到了企业认可的标准。     

基于神经网络的钢筋计数方法研究

为了提高钢筋的计数准确率和效率,综合运用图像处理技术和神经网络技术,实现对钢筋的识别和计数.对获取的钢筋原始图像进行数字图像处理,得到感兴趣的部分即钢筋的轮廓;计算单根钢筋轮廓的宽度、高度、面积和打捆钢筋瞧总面积4个特征量;将这4个特征量作为神经网络的输入,训练网络识别钢筋并计数.仿真实验验证了这种方法的可行性和有效性.     

CATALOG: a system for detection and rendering of internal log defects using computer tomography

This paper describes the design and implementation of a machine vision system CATALOG for detection and classification of some important internal defects in hardwood logs via analysis of computer axial tomography (CT or CAT) images. The defect identification and classification in CATALOG consists of two phases. The first phase comprises of the segmentation of a single CT image slice, which results in the extraction of 2D defect-like regions from the CT image slice. The second phase comprises of the correlation of the 2D defect-like regions across CT image slices in order to establish 3D support. The segmentation algorithm for a single CT image is a complex form of multiple-value thresholding that exploits both, the prior knowledge of the wood structure within the log and the gray-level characteristics of the image. The algorithm for extraction of 2D defect-like regions in a single CT image first locates the pith of the log cross section, groups the pixels in the segmented image on the basis of their connectivity and classifies each 2D region as either a defect-like region or a defect-free region using shape, orientation and morphological features. Each 2D defect-like region is classified as a defect or non-defect via correlation across corresponding 2D defect-like regions in neighboring CT image slices. The 2D defect-like regions with adequate 3D support are labeled as true defects. The current version of CATALOG is capable of 3D reconstruction and rendering of the log and its internal defects from the individual CT image slices. CATALOG is also capable of simulation and rendering of key machining operations such as sawing and veneering on the 3D reconstructions of the logs. The current version of CATALOG is intended as a decision aid for sawyers and machinists in lumber mills and also as an interactive training tool for novice sawyers and machinists. Received: 1 August 1997 / Accepted: 25 August 1999     

搜索日志中中文人名自动识别

搜索日志中人名识别一直是日志挖掘中的一个重点和难点,其结果好坏直接关系搜索引擎的检索效率和准确率。由于分析了长文本中人名识别方法在搜索日志中使用存在很多困难与不足,因而该文提出了一种在搜索日志中识别中文人名的方法。该方法将搜索日志中人名内部用字的概率特征引入条件随机场,再根据搜索日志的特点计算人名可信度提取搜索日志中的中文人名。在搜狗查询日志上进行实验,正确率平均达到了81.97%、召回率平均达到了85.81%,综合指标F值平均达到了83.79%。     

基于词性标注的启发式在线日志解析方法

为了解决现有启发式日志解析方法中日志特征表示区分能力不足导致解析精度低、泛化差的问题,提出了一种启发式在线日志解析方法PosParser。该方法使用来源于触发词概念的功能词序列作为特征表示,包含解决复杂日志易过度解析问题的两阶段检测方法和处理变长参数日志的后处理流程。PosParser在16个真实日志数据集上取得了0.952的平均解析准确率,证明了功能词序列具有良好区分性、PosParser有良好的解析效果和鲁棒性。     

Identification of stratigraphic formation interfaces using wavelet and Fourier transforms

The purpose of this study was to identify the formation interfaces from geophysical well log data using the wavelet transform, and a combination of the wavelet transform and the Fourier transform methods. In the wavelet transform method, the identification of formation interfaces is based on the wavelet coefficients from the wavelet transform of spontaneous potential (SP) log and gamma ray (GR) log data. In the combination of the wavelet transform and the Fourier transform methods, the wavelet transform, spectrum analysis, and logarithmic transform of well logs were applied to the SP and GR log data successively to obtain clear signals for identifying the stratigraphic formation interface. In this study, a set of ideal log data was first created and analyzed to test the validity of the developed procedures. In analyzing the SP and GR logs from a field, both the wavelet transform method and conventional well log analysis showed similar results. The results from a combination of the wavelet transform and the Fourier transform methods, however, were better than those from the wavelet transform method and the conventional well log analysis.     

基于轮廓特征点的高精度识别目标方法

为了实现高成功率、高精度和快速识别跟踪目标,提出对基于轮廓特征点的目标精确识别方法。在识别过程中采用了轮廓提取和多边形拟合算法自动搜寻到图像中要识别和跟踪的目标,同时对目标物轮廓的多边形角点进行亚像素分辨率的定位,从而可以利用目标轮廓角点的精确定位来实现对多边形目标的识别与跟踪。试验结果以及特征点亚像素算法分辨率的分析表明,采用这种自动识别与跟踪目标的方法,其精度可以达到0．02像素。     

Incorporating Performance Testing in Test-Driven Development

Our performance-testing approach required manually inspecting the performance logs. During the project's development, JUnit-based performance testing tools, such as JUnitPerf, weren't available. Such tools provide better visibility of performance problems than manual inspection of performance logs. Although we believe manual inspection of performance trends is necessary, specifying the bottom-line performance in assert-based test cases can complement the use of performance log files, making the TFP testing results more visible to the developers. We're investigating the design of assert-based performance testing to improve the TFP process. Another direction of future work is automatic performance test generation. In this project, we relied on the performance architect's experience to identify the execution paths and measurement points for performance testing. We can derive this crucial information for performance testing from the performance requirements and system design. We plan to find guidelines for specifications of performance requirements and system design to make the automation possible     

Web日志挖掘的用户识别算法研究

Web日志挖掘是将数据挖掘技术应用到Web服务器的日志中,发现Web用户的行为模式,以便进一步改善网站结构或为用户提供个性化的服务。文中探讨了Web日志挖掘中的用户识别算法,提出了一种多重约束条件的用户识别算法。     

CPL+: An improved approach for evaluating the local completeness of event logs

Process mining aims at deriving order relations between tasks recorded by event logs in order to construct their corresponding process models. The quality of the results is not only determined by the mining algorithm being used, but also by the quality of the provided event logs. As a criterion of log quality, completeness measures the magnitude of information for process mining covered by an event log. In this paper, we focus on the evaluation of the local completeness of an event log. In particular, we consider the direct succession (DS) relations between the tasks of a business process. Based on our previous work, an improved approach called CPL+ is proposed in this paper. Experiments show that the proposed CPL+ works better than other approaches, on event logs that contain a small amount of traces. Finally, by further investigating CPL+, we also found that the more distinct DSs observed in an event log, the lower the local completeness of the log is.