Real-time pedestrian detection using SVM and AdaBoost

Abstract

This project presents an application of pedestrian detection system to reduce the number and severity of vehicle-pedestrian accident by active safety vehicle. It is important to detect pedestrian efficiently and accurately in many computer vision applications, such as intelligent transportation systems and safety driving assistant systems. In this system, we are presenting a pedestrian detection method based on images. We are using Ada-Boost algorithm and cascading methods to segment pedestrian candidates from image. To confirm whether each candidate is pedestrian or not a pedestrian. Recognizing classifier is skilled with support vector machine (SVM). We are giving input features used for SVM training are mined from both the sample gray images and edge images to the system.