本页对应的英文页面已更新,但尚未翻译。 若要查看最新内容,请点击此处访问英文页面。

常见几何变换

使用 imwarp 工作流执行常见几何变换

几何变换将输出图像中的像素坐标映射到输入图像中的坐标。然后,映射过程根据输入图像对输出像素进行插值。

使用这些函数可执行一般的二维、三维和 N 维几何变换。要执行二维或三维几何变换,请首先创建一个几何变换对象,用来存储变换信息。然后,将要变换的图像和几何变换对象传递给 imwarp 函数。

函数

imwarpApply geometric transformation to image
affineOutputViewCreate output view for warping images
fitgeotrans对控制点对组进行几何变换拟合
findboundsFind output bounds for spatial transformation
fliptformFlip input and output roles of spatial transformation structure
makeresamplerCreate resampling structure
maketformCreate spatial transformation structure (TFORM)
tformarrayApply spatial transformation to N-D array
tformfwdApply forward spatial transformation
tforminvApply inverse spatial transformation

对象

全部展开

WarperApply same geometric transformation to many images efficiently
imref2dReference 2-D image to world coordinates
imref3dReference 3-D image to world coordinates
affine2d二维仿射几何变换
affine3d 3-D affine geometric transformation
projective2d 2-D projective geometric transformation
geometricTransform2d2-D geometric transformation object
geometricTransform3d3-D geometric transformation object
PiecewiseLinearTransformation2D2-D piecewise linear geometric transformation
PolynomialTransformation2D2-D polynomial geometric transformation
LocalWeightedMeanTransformation2D2-D local weighted mean geometric transformation

主题

几何变换

2-D and 3-D Geometric Transformation Process Overview

To perform a general geometric transformation of a 2-D or 3-D image, first define the parameters of the transformation, then warp the image.

几何变换的矩阵表示

仿射变换和投影变换以矩阵表示。您可以使用矩阵运算来执行图像的全局变换。

N-Dimensional Spatial Transformations

You can create custom geometric transformations to process images of arbitrary dimension, or to change the dimensionality of the output image from the input image.

Specify Fill Values in Geometric Transformation Output

This example shows how to specify the color of blank space in the image after a geometric transformation.

空间参照

图像坐标系

了解如何使用像素索引和空间坐标来表示图像位置。

Define World Coordinates Using Spatial Referencing

Spatial referencing objects encode the relationship between the image extent in intrinsic coordinates, the image extent in world coordinates, and the image resolution.

Define World Coordinates Using XData and YData Properties

The XData and YData image properties are two-element vectors that control the range of coordinates spanned by the image.

特色示例