Vision-based pedestrian detection and estimation with a blind corner camera

Scholarly Research Paper from the year 2006 in the subject Electrotechnology, grade: 1,0, University Karlsruhe (TH), language: English, abstract: Avoiding collision accidents is becoming more and more an important topic in the researchfield of driver assistant systems. Especially for vision-based detection systems there arevarious approaches, which are built upon many different methods.This thesis deals with the avoidance of pedestrian accidents, caused by Blind Cornerview problems. The presented approach comprises a pedestrian detection subsystem, whichis part of a large camera system framework covering observation of the car environment.Based on a Blind Corner Camera and a neural network classification method, research inthis thesis is focused on two aspects: detection improvement and danger level estimation.Since vision-based classification methods usually are still not able to yield perfect results,because of the complexity of this task, the detection result has to be improved bypreprocessing and post processing. In this work, first, effects of image enhancementmethods on detection are tested as preprocessing methods and, secondly, a new approachfor a simple tracking and estimation strategy is presented, which improves detection in theway of a post processing method. Finally, information from tracking and prediction is usedto estimate a danger level for pedestrians, which provides information about how collisionpronethe current situations is.