Now we have many way to get it , but the fiber scale in this paper is not enough obvious to distinguish it , we get the edge man made in photoshop . then we fill black in it so as to convert a image which only have two colors : black and white . the area can be calculated by counting dot of black 但是由于本课题的鳞片边缘灰度与背景相差不大，用传统的边缘检测方法检测不出来，所以使用photoshop人为提取法，提取出边缘后内部填充黑色，使物体与背景有较强的对比作用，选取适当的阈值进行图像分割，使所有灰度值大于或等于某阈值的像素都被判属于物体。
The extracting methods of flavonoids from plants including traditional extracting methods and new methods such as microwave extraction , supercritical extraction , ultrasonic extraction , enzyme hydrolysis extraction , and semi - bionic extraction were reviewed in this paper in order to offer reference to the research , development and application of flavonoids 摘要综述了黄酮类化合物的提取方法，包括传统的提取方法和新型提取技术如超声提取、微波提取、超临界流体提取、酶法提取、半仿生提取法，旨在为黄酮类化合物的研究、开发、应用提供参考。
The histological descriptions of p . armandi franch in chinese herbal compendium ( zhong hua ben cao ) are different form our observations and the descriptions in chinese flora and sichuan flora in the aspects of the shape of the transverse section and the number of vascular bundles . by means of gc - ms , qualitative and quantitative analysis of the essential oils obtained from different species , different production areas and different extracting methods were carried out . the chemical components and their contents of the essential oils vary by the species , production area and extracting method 通过对不同品种(马尾松、云南松、华山松) 、马尾松不同产地(万源、夹江、资中、泸州、威远、康定) 、不同提取方法(水蒸气蒸馏法、超临界提取法)的松叶挥发油成分的gc - ms分析比较，表明挥发油成分的组成和含量有差异，且品种、产地、提取工艺均对松叶的挥发油成分有影响。
Thirdly , decrease the difficulty of the character recognition through studying the method of character preprocessing ; at last , some effective methods of feature extraction , such as the one based on image transform with the invariant features which keep stable while the picture move , rotate and zoom in or out , and the other focused on the shape of character and features of direction 三、深入探讨了字符识别的预处理方法，减少字符识别时的运算量和难度。四、讨论了三种特征向量的提取方法，一种是基于平移、旋转和尺度不变性的图像变换法，一种是计算中心矩的方法，再就是强调字符形状和结构的方向特征量提取法，实验中通过最小距离法进行字符的分类识别，匹配时间少，识别率有所提高。
Thus , the moving target tracking is a lot subject combining with knowledge of the digital signal processing , the image processing , pattern recognition , artificial intelligence and etc . at first , the automatic detection method of moving target is studied , and the background image ascertaining method based on the character matching is proposed 因此,运动目标跟踪是一个多学科交叉的研究方向,它融合了数字信号处理、图像处理、模式识别的知识。本文首先研究了运动目标的自动检测方法,提出了基于特征匹配的背景提取法,并利用背景差法实现了运动目标检测和提取。
Theories of linear prediction approach , matrix prediction approach and state space approach are introduced . it can be shown that linear prediction approach is one kind of matrix prediction approach , and the latter is equivalent to the second methods of state space approach , so the prony method , the kt ( svd - prony ) method , and the mpm method are connected in state space 本文详细介绍了极点提取法中的线性预测法,矩阵预测法及状态空间法的原理,表明线性预测法是矩阵预测法的一个特例,而矩阵预测法与状态空间法的算法二等价,从而使prony方法, kt ( svd - prony )方法, mpm方法在状态空间意义下得到统一。
This study is backed by the research project of stimulus onset asynchrony ( soa ) detection , a newly devised multisensory erp experimental method . on the basis of spatiotemporal pca and existing temporal component extraction method , a spatial component extraction method was designed and materialized . detailed algorithm description , software development introduction and application example discussion were all presented 本研究以一种新型的多感知erp实验方法?异步刺激检测法( soadetection )的研究项目为背景，以时空pca与现有的时间成分提取法为基础，进行了空间成分提取法的算法设计，并做了软件开发及应用分析。
