M.ain Content

eStimateGeometricTransform.

从匹配点对估计几何变换

在页面上崩溃

句法

tform = estimationgeometrictransform（matchedpoints1，matchedpoints2，transformitype）
[tform，Inlierpoints1，Inlierpoints2] = eStimateGeometricTransform（MatchedPoints1，MatchedPoints2，Transformitype）
[___那status] = estimateGeometricTransform(matchedPoints1,matchedPoints2,transformType)
[___] = eStimateGeometricTransform（MatchedPoints1，matchedpoints2，transformitype，名称，值）

Description

例子

tform= eStisimeGeometricTransform（matchedPoints1那matchedPoints2那transformType的）返回2-D几何变换对象，tform。Thetform对象映射inliersmatchedPoints1对替代者matchedPoints2。

该功能不包括使用M估计的样本共识（MSAC）算法的异常值。MSAC算法是随机样本共识（RANSAC）算法的变体。Results may not be identical between runs because of the randomized nature of the MSAC algorithm.

笔记

This function will be removed in a future release. Use theeStimateGeometricTransform.2DoreStimateGeometricTransform.3.D函数提供更多的功能。

例子

[tform那inlierpoints1那inlierpoints2] = eStisimeGeometricTransform（matchedPoints1那matchedPoints2那transformType的）返回相应的Inlier点inlierpoints1andinlierpoints2。

例子

[___那status] = eStisimeGeometricTransform（matchedPoints1那matchedPoints2那transformType的）返回状态代码0.那1， 要么2。如果你不请求statuscode output, the function returns an error for conditions that cannot produce results.

[___] = eStisimeGeometricTransform（matchedPoints1那matchedPoints2那transformType那姓名那Value的）uses additional options specified by one or more姓名那Valuearguments.

Examples

collapse all

Open Live Script

Read an image and display it.

原始= imread（'cameraman.tif'）;imshow(original) title('Base Image'的）

扭曲并显示变换的图像。

distorted = imresize(original,0.7); distorted = imrotate(distorted,31); figure imshow(distorted) title('转换图像'的）

Detect and extract features from the original and the transformed images.

ptsoriginal =探测器（原始）;ptsdistorted =检测（扭曲）;[特征素，validptsoriginal] =提取物（原始，冒号）;[具有截图，validptsdistorted] =提取物（扭曲，ptorted，ptordistored）;

M.atch and display features between the images.

index_pairs = matchfeatures（特征逻辑，特点）;matchedptsoriginal = validptsoriginal（index_pairs（：，1））;matchedptsdistored = validptsdistorted（index_pairs（:,2））;Figure ShowMatchedFeatures（原始，扭曲，Matchedptsoriginal，MatchedptsDistorted）标题（'Matched SURF Points With Outliers'）;

Exclude the outliers, estimate the transformation matrix, and display the results.

[tform，inlieridx] = estimateGeometricTransform2D（MatchedPtsDistorted，MatchedPtSoriginal，'similarity'）;InLierptsDistorted = matchedptsdistorted（inlieridx，:);Inlierptsoriginal = matchedptsoriginal（Inlieridx，:);Figure ShowMatchedFeatures（原始，扭曲，inlierptsoriginal，InLierptsdistorted）标题（'Matched Inlier Points'的）

Use the estimated transformation to recover and display the original image from the distorted image.

outputView = imref2d(size(original)); Ir = imwarp(distorted,tform,'OutputView'那outputView); figure imshow(Ir); title('Recovered Image'）;

Input Arguments

collapse all

来自image 1的匹配点，指定为aKAZEPoints那cornerPoints目的那截图那Surfpoints.目的那MSERREGIONS.目的那ORBPoints目的那Brikspoints.or anM.-2矩阵[x，y]坐标。该功能不包括使用M估计的样本共识（MSAC）算法的异常值。MSAC算法是随机样本共识（RANSAC）算法的变体。

M.atched points from image 2, specified as either aKAZEPoints那cornerPoints目的那截图那Surfpoints.目的那MSERREGIONS.目的那ORBPoints目的那Brikspoints.or anM.-2矩阵[x，y]坐标。该功能不包括使用M估计的样本共识（MSAC）算法的异常值。MSAC算法是随机样本共识（RANSAC）算法的变体。

变换类型那specified as one of three character strings. You can set the transform type to either'similarity'那'仿射'， 要么'投影'。匹配成对的数量越大，估计变换的准确性越大。每个变换类型的最小匹配点数数：

变换类型 M.inimum Number of Matched Pairs of Points
'similarity' 2
'仿射' 3.
'投影' 4.

数据类型：char

名称 - 值参数

指定可选的逗号分离对姓名那Valuearguments.姓名是参数名称和Valueis the corresponding value.姓名must appear inside quotes. You can specify several name and value pair arguments in any order as姓名1那Value1,...,NameN,ValueN。

Example:'Confidence'那99将置信度值设置为99。

M.aximum number of random trials for finding the inliers, specified as the comma-separated pair consisting of 'maxnumtrial'和一个正整数标量。增加此值以额外计算为代价提高了结果的稳健性。

数据类型：单身的|double|INT8.|int16|int32|int64|uint8.|uint16|UINT32|UINT64

信心of finding the maximum number of inliers, specified as the comma-separated pair consisting of '信心' and a percentage numeric scalar in the range (0 100). Increasing this value improves the robustness of the results at the expense of additional computations.

数据类型：单身的|double|INT8.|int16|int32|int64|uint8.|uint16|UINT32|UINT64

M.aximum distance in pixels, from a point to the projection of its corresponding point, specified as the comma-separated pair consisting of 'Maxdistance.'和一个正数标量。相应的投影基于估计的变换。

数据类型：单身的|double|INT8.|int16|int32|int64|uint8.|uint16|UINT32|UINT64

输出参数

collapse all

Geometric transformation, returned as either anAffine2d.目的or aprojective2d目的。

The returned geometric transformation matrix maps the inliers inmatchedPoints1对替代者matchedPoints2。When you set thetransformType输入到任一个'similarity'or'仿射'，函数返回一个Affine2d.目的。Otherwise, it returns aprojective2d目的。

Status code, returned as the value0.那1， 要么2。

status Description
0. No error.
1 matchedPoints1andmatchedPoints2输入不包含足够的点。
2 发现没有足够的恋人。

如果你不请求status代码输出，函数将为无法生成结果的两个条件抛出错误。

数据类型：double

Image 1中的Inlier点，返回与输入匹配点相同的类型。

Inlier points in image 2, returned as the same type as the input matching points.

References

[1] Hartley, R., and A. Zisserman, "Multiple View Geometry in Computer Vision,"剑桥大学出版社那20.0.3.。

[2] Torr, P. H. S., and A. Zisserman, "MLESAC: A New Robust Estimator with Application to Estimating Image Geometry,"计算机视觉和图像理解那20.0.0.。

扩展能力

也可以看看

对象

职能

话题

Introduced in R2013a

计算机视觉工具箱文档

金宝app

Introducing Deep Learning with MATLAB

Introducing Deep Learning with MATLAB

Download ebook