Warp 2d

Modules demonstrating how apply a 2d warping transformation on an image. More...

Modules demonstrating how apply a 2d warping transformation on an image.

This sample illustrate how to correct perspective effects on a plan.

We will define (at least) four input points in input image which will be used to correct perspective effects :

We start by importing all necessary libraries:

import os
import PyIPSDK
import PyIPSDK.IPSDKIPLGeometricTransform as gtrans

Then we load input image :

# opening of input image
imagesSamplePath = PyIPSDK.getIPSDKDirectory(PyIPSDK.eInternalDirectory.eID_Images)
inputImgPath = os.path.join(imagesSamplePath,  "IPSDK_flyer1.tif")
inImg = PyIPSDK.loadTiffImageFile(inputImgPath)

We then select input image points for which we define target output image coordinates :

# we define a 2d mapping between 4 points of input image and target points
# for output image
# on output we want to get a rectangular view
targetWidth = 400
targetHeight = 550
inPtColl = [[327, 83],
            [697, 216],
            [459, 763],
            [29, 552]]
outPtColl = [[100, 100],
             [100+targetWidth, 100],
             [100+targetWidth, 100+targetHeight],
             [100, 100+targetHeight]]

These data allow us to compute associated homography 2d transform :

# we compute associated homography
transform, res = PyIPSDK.homographyTransform2dSimpleEstimation(inPtColl, outPtColl)

For more informations about parametric estimation features, please see Parametric estimation.

Computed transform can then be used to warp input image to output image :

# application of warping transform
outImg = gtrans.warp2dImg(inImg, transform)

See also

Warping 2d algorithm for more informations

Note

user could define and pass output image to warping function to define a target output image size.

At last we save output image :

# save generated image
tmpPath = PyIPSDK.getIPSDKDefaultDirectory(PyIPSDK.eDefaultExternalDirectory.eDED_Tmp)
outputImgPath = os.path.join(tmpPath, "warp2dImg.tif")
print("Writing result in file : " + outputImgPath)
PyIPSDK.saveTiffImageFile(outputImgPath, outImg)

See the full source listing