Image processing mini-projects I'm working on using OpenCV.
Detect and track the lane lines on a highway. Also performed on video.
Feature matching method comparisons using ORB keypoints/descriptors:
Stitching two images from my living room using SIFT.
Idea for app that could estimate trajectory for going off a ski jump by having user take a picture of the jump and select 3 points, 1) starting location 2) bottom of jump 3) top of jump. Users could build a jump and visually determine if the jump is big enough/how high to drop from/if they'll clear the landing by taking a picture with their phone. Best performance when jump viewed directly from the side, can be viewed from either side.
Edge detection comparisons between color space representations:
CLAHE enhancement for improved visibility under flat light conditions:
Brightest point:
Varying size, colors and overlaps:
Fusing images with different exposure levels (Auto Exposure Bracketing / High Dynamic Range):
