粗检测采用基于标准行的灰度匹配法来判断疵点是否存在，细检测对疵点进行定位和聚类。
Rough detection uses standards-based matching method to determine the existence of defects, fine detection includes positioning and clustering.
灰度匹配是利用图像的灰度对两幅图像进行立体匹配，以相关函数为判断依据对图像进行立体匹配。
The paper processes gray match with the image gray and on the basis of correlative function. The paper processes feature stereo match with zero crossing.
首先利用灰度投影法得到相邻两幅图像的平移量，在此基础上再利用块匹配算法进行分块处理。
First, we calculate the translation between two images by gray projection algorithm, then we use the block-matching algorithm in the divided blocks.
本文提出了一种基于灰度曲面匹配的人耳识别方法。
An ear recognition method based on gray surface matching has been proposed.
然后，根据影像的长度变形、方向变形、灰度变形及双向一致性条件，计算匹配支持度；
Then the matching support can be calculated according to length distortion, direction distortion, gray distortion, and double faces constraint.
然后根据这种算法进行垂直边界匹配来从输入的灰度图像中识别车牌。
Then a vertical edge matching based algorithm recognizes license-plate from input gray-scale image.
首先在序列的两幅图像中检测兴趣点，并运用灰度差分不变量进行两点集之间的匹配。
Firstly, we detect the points of interest in a pair of images in image sequence, and then match the point set using the gray value differential invariants.
针对图像灰度变化和遮挡问题，提出一种增量符号相关的景象匹配方法。
A scene matching approach based on increment sign correlation is proposed to solve the problem of gray change or occlusion in image.
在获得了兴趣点之后，利用兴趣点处的局部灰度差分不变量进行序列图像的点特征匹配。
Once the interest points were detected, the image matching process in an image sequence is performed using local gray-value differential invariants.
在使用模板匹配方法检测织物瑕疵的过程中，通过实时采集、分析织物的灰度图像，获得织物纹理的统计信息，并从中提取出正常纹理的特征。
In the detection of fabric flaw based on template matching, statistical information of natural and normal texture of the fabric can be extracted via collecting and analyzing the gray image.
针对图像匹配受几何变形的影响，提出了一种基于相邻像素间的灰度差异来提取图像多像素边缘的方法。
To reduce the influence of geometrical deformation on image matching, we put forward a method which can detect multi-pixels edge of image.
本文的方法以图像分割之后的区域为匹配基元，灰度相差较大的前景和背景区域分割之后将归属于不同区域，所以能较好的解决该问题。
After divided, the foreground and background regions which have biggish gray difference will belong to different regions, therefore the method can solve the problem mentioned before.
针对此情况提出了采用一维灰度投影的模板匹配方法。
Under this circumstance, the author puts forward the template-matching algorithm based on one-dimensional gray projection.
利用最小外接矩形和灰度信息作为各目标的特征向量，匹配各目标对象，以求取各目标运动速度。
The outer rectangular bounds and tonal-gradation information are employed to be eigenvectors to match respective targets, so as to get the speed of them.
灰度失真、噪声、几何失真等误差因数会影响匹配性能，本文给出了相对应的处理方法。
The error factors such as intensity distortion, noise and geometry distortion will affect matching performance, so the paper gives the opposite processing methods.
通过对制导图的灰度最小化预处理，提高了匹配算法的抗灰度畸变能力。
Through grayscale minimization preprocessing on images for guidance, the anti grayscale ability of matching algorithm is improved.
然后引入色彩相似性约束将灰度空间的匹配和彩色空间的匹配有机的结合起来，大大提高了匹配的精度。
Then, by adding constraint of color similarity, the grey space matching was combined with color space matching suitably and the improved SSD method could use the color information reasonably.
分析了传统的基于灰度和基于特征的匹配算法，提出了一种基于边缘特征和相关系数的新匹配算法。
A hybridized matching algorithm is proposed based on edge feature and correlation coefficient method after analyzing traditional grey-based algorithms and feature-based algorithms.
该方法采用事先通过标定的图像，对自然标志(特征点)的提取和基于灰度向量匹配来完成绝对位移的测量。
By using calibrated image, it can measure plane absolute displacement of worktable automatically via feature point extraction and gray vector matching.
由于画像与照片的产生机理和信息表达方式不同，两者之间存在较大的几何形变及纹理和灰度差异，若直接采用画像与照片之间的匹配程度进行识别势必产生很大误差。
Due to large differences between a photo-sketch pair caused by different generation mechanism and information expression manner, it is impossible to recognize by simply matching the photo-sketch pair.
另一种方法是分别得到横竖条码在水平和垂直方向的一维灰度投影曲线，根据它们的相关性判断两幅图像是否匹配。
The other method uses 1-D cyclostyle matching algorithm to obtain horizontal and erect gray scale projection curves of two list images, then compares with them if its not homology.
利用兴趣点进行序列图像匹配可以利用兴趣点及其邻域的灰度信息和兴趣点集的几何信息。
The information available for image matching includes the gray-level of or near interest points and the geometrical relation of interest points.
算法采用灰度投影获取全局运动，用网格检测和灰度切分图精确匹配的方法获取更准确的运动数据。
The image motion is obtained by the gray project and then the motion data is accurately detected by the grid detection and bit plane match method.
本文研究的结论对于合理地设计画面灰度标识使之符合人机匹配原则具有意义。
The conclusions given in this paper are significant in designing gray level symbols of a pattern to match the visual needs in the man-machine system.
在立体匹配时，将灰度相关系数小于最大灰度相关系数一定范围内的特征点作为灰度相关复峰初始匹配特征点集合。
During stereo matching, the feature points whose correlation coefficients are within certain range less than the maximal coefficient can be taken as an original matching feature point set.
灰度相关直接用图像灰度进行匹配，没有考虑场景物体之间的空间和结构相关性，使得结果对结构、光照、纹理和畸变十分敏感。
Matching is performed directly with the image intensity in the area-based correlation without counting the space and structure correlation between objects.
它包括二维匹配滤波预处理以增强血管的灰度，以及用灰度-梯度共生矩阵的最大熵阈值化方法。
Entropy thresholding method is an automatic technique for thresholding of digital images based on gray level-gradient co-occurrence matrix and the maximum entropy principle.
基于灰度的数字图像立体匹配算法的中心问题是选择一个适当尺寸的窗口。
A central problem in stereo matching by computing correlation or sum of squared differences (SSD) lies in selecting an appropriate window size.
对两幅图像进行预处理后进行特征点的匹配，以特征点的灰度相关性作为判据完成左右图像的特征点匹配。
After processing two images, we can match the character points. The left and right image character points had matched by their gray relativity.
本文的研究分为两个部分：一是基于灰度的图像匹配算法研究，二是运动目标的跟踪算法及仿真研究。
The research of this paper is divided into two parts. One is intersity-based matching algorithms, and the other is tracking algorithm and simulation of moving objects.
应用推荐