Computer Vision for Student Competitions: Object Detection - Part 1
Updated 12 Oct 2017
Learn to detect objects using binary classifiers; template matching, histogram of gradients (HOG), and cascade object detection.
You’ll learn how Template Matching works and how to use it. The concepts behind HOG will be taught to prepare you for the Cascade Object Detector. The Cascade Object Detector is a robust detector which provides the option to use Haar, Local Binary Patterns (LBP), and HOG to detect objects within an image.
Template Matching and Cascade Object Detection are used to detect objects in an image that are aspect ratio and orientation invariant. Template Matching has an additional limitation that the object must be scale invariant. Both methods are useful for determining if an object is located within an image, and if so, where the object is located within the image.
Using these methods, teams in the AUVSI Foundation competitions should be able to perform a variety target identification tasks.
Images from bike and non-bike folders were obtained from the object detection image database created by Dr. Axel Pinz from the Graz University of Technology. www.tugraz.at/en/institutes/emt/personal-pages/pinz/
MathWorks Student Competitions Team (2021). Computer Vision for Student Competitions: Object Detection - Part 1 (https://github.com/mathworks/auvsi-cv-object-1), GitHub. Retrieved .
MATLAB Release Compatibility
Platform CompatibilityWindows macOS Linux
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!Start Hunting!