OpenCV image stitching analysis

• Main Author: Kok, Chuan Yong
• Other Authors: Lee Bu Sung, Francis
• Format: Final Year Project
• Language: English
• Published: 2018
• Subjects: DRNTU::Engineering
• Online Access: http://hdl.handle.net/10356/73901
• Institution: Nanyang Technological University

Computer vision software has been widely used in the area of Photography creating an amazing effect. One of the tools that has been developed and widely used in the digital camera is the panoramic image, where multiple images are “stitched” together. In this project, the analysis was carried out on the open source library OpenCV and two of its feature detectors ORB and AKAZE. Three programs were built or updated from the old version of OpenCV to the newer version to carry out the experiments. One is the ABS-DIFF program which calculated the absolute difference between two images. The program is used to measure the difference between the original image and the “stitch” image. The second draw matches program was also developed to highlight the “matched” points when we stitch two images together. This program helps us to understand the stitching software better as it shows the matching points of the reference and target image side by side. The third program is an image stitcher which was modified and updated from an old version to verify the stitching process and runtime. A number of experiments were carried out to compare ORB and AKAZE. It is concluded that ORB has a better performance in terms of computation time requirement and AKAZE has better performance in the cumulative absolute difference value.