Image Stacking Save

Automatic Image Stacking in OpenCV

Project README

Automatic Image Stacking in OpenCV

Image stacking is a technique in digital photograpy where multiple photos from the same scene are "stacked" together in order to reduce image noise caused by a high ISO values. This allows for creating a low noise image at low light conditions. In order to fuse multiple images together the pixels in all photos needs to aligned with each other, which is usually not possible when taking images without a tripod. This python script shows an example of stack multiple images together by first automatically alligning and then calculating a the fused images as a pixel-wise mean of the alligned images.

To see the power of image stacking, check out the before and after images below. Left is a single image taken at in low light, and right is 8 images stacked from the same scene stacked toghether. The more images you have the more the noise can be reduced.

alt text

In order to allign the images the perspective transforms between the first images and every other images are to be estimated. The perspective transform can be estimated either using ORB Keypoint matching (faster) or ECC maximization (more precise).

alt text


  • Python 2 or 3
  • OpenCV 3.0

Example Usage:

ORB method (faster):

  python images/ result.jpg --method ORB

alt text

ECC method (more precise):

  python images/ result.jpg --method ECC

alt text

Open Source Agenda is not affiliated with "Image Stacking" Project. README Source: maitek/image_stacking
Open Issues
Last Commit
2 years ago

Open Source Agenda Badge

Open Source Agenda Rating