Pedestrian Detection Using PCA
December 2024
At a glance
Skills: MATLAB, computer vision, principal component analysis
What: Created a pedestrian detection algorithm using Principal Component Analysis (PCA).
How: Implemented algorithm in MATLAB to determine whether a pedestrian was close by in an image with a 78% accuracy rate. Utilized PCA to reduce the dimensionality of the input images, increasing computational efficiency.
Why: Pedestrian detection for vehicle safety features must function in resource constrained systems, making efficient algorithms crucial.
Details
Research poster presenting our methodology and results
For my final project in Quantitative Engineering Analysis 1 (Linear Algebra) I worked with a partner to create a pedestrian detection algorithm using principal component analysis (PCA).
Our algorithm, implemented in MATLAB, determined whether a pedestrian was close by in an image with a 78% accuracy rate. Our model was trained and tested on a subset of images from the Caltech pedestrian detection dataset.