This repo contains the lab code in the 2021 Computer Vision for UAV Autopolot course.
In this lab, image was rotated, flip and zoom from scratch. For the zoom fuction, nearest and bilinear interpolation methods were tested.
-Original image:
-Nearest interpolation result:
-Bilinear interpolation result:
In this lab, histogram equalization was established. Besides, Otsu algorithm was consturted form scratch.
-Input image
-Result Otsu image
In this lab, people in the video were labeled by connected component method. In addition, two-pass method was performed.
-Result video
In this lab, we trained yolo to identify two different characters and position in the frame. The result is shown in the following video.
-Result video
In this lab, we compare the different method of image charateristic extraction.
-Result:
In the final, ORB-SLAM and Colmap were used to modeling the environmnet.
The blue line was Colmap result and the orange dots was ORB-SLAM result.
-Result: (rmse = 0.8131214164989768 )
