ORB feature based image stitching for UAV images

Image stitching is the process of stitching multiple images with overlapping areas to produce a panoramic or high resolution image. Unmanned Aerial Vehicle (UAV) can be widely used to study the terrain. Each time a UAV sweeps through an area it takes images on different sections of the regions. An image stitching software is used to stitch these images into a single image based on overlapping regions. This is a python script for doing the same. Here I used ORB features to identify each feature in an image and using that the images are stitched together. ORB is a pre-built function available on openCV. The data for the program are cropped regions of a UAV image, stored in a folder. The path of the folder is given as input. Program iterate through the folder and joins image based on common feature.

Input

Output

How to run the code

1. Clone the repository in your machine
2. Install requirements.txt using pip
3. Add your data on same folder
4. Run UAV-image-stitching.py