It is aimed to obtain a final panorama image that will combine the subimages provided using key point identification methods (SIFT/SURF and ORB) and include all the scenes in the subimages.
Since the method I used to create a panorama image from subimages is a bit long, you can export the dataset from a small number of images.
- python3 (or higher)
- opencv 3 (or higher)
You will need to install some package using pip
:
- numpy
- matplotlib
$ python main.py <feature extraction method name(SIFT or ORB)>
$ python main.py SIFT $ python main.py ORB $ python main.py // SIFT is selected manually
The dataset folder must be in the same directory as main.py
The program will output:
- Showing every feature extraction, matched and Registration images by using matplotlib. If you don't want to see it. Comment lines 45, 50, 67, 94 in merge.py.
- A panoma stitched images
Stitched_Panorama_Full_{pano_name}.png
I test my code in Window11. It should work fine in these system.