↓ Skip to main content
Jacob Keller

Pedestrian Detection Using PCA

·154 words·1 min
School Project

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
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.

Resources
#