Reinhard Janssen

A real-time traffic sign recognition system

The ability of recognising traffic signs in a road traffic scenario is an important feature of the Daimler-Benz autonomous vehicle VITA II. This real-time vision-based traffic sign recognition system has been developed by Daimler-Benz in the European research project PROMETHEUS. In this paper we focus on the overall system design, the real-time implementation, and field test evaluation. The software architecture of the system integrates three hierarchical levels of data processing. On each level the specific tasks are isolated. The lowest level comprises specialists for colour, shape and pictogram analysis; they perform the iconic to symbolic data transformation. On the highest level the administration processes organise data flow as a double bottom-up and top-down mechanism to dynamically interpret the image sequence. A hybrid parallel machine was designed for running the traffic sign recognition system in real time on a transputer network coupled to powerPC processors.

An adaptive system for traffic sign recognition

Traffic sign recognition is a primary goal of almost all road environment understanding systems. A vision system for traffic sign recognition was developed by Daimler-Benz Research Center Ulm. The two main modules of the system are detection and verification (recognition). Here regions of possible traffic signs in a color image sequence are first detected before each of them is verified and recognized. In this paper the authors pay attention to the verification and recognition process. The authors present an adaptive approach and emphasize the importance of the adaptability to various road and traffic sign environments. The authors utilize a distance-weighted k-nearest-neighbor classifier for traffic sign recognition and show its equivalence to the kind of radial basis function networks which can be easily integrated into chips. The authors also present a way to evaluate the uncertainty of recognized traffic signs and demonstrate their approach using real images.

Hybrid Approach For Traffic Sign Recognition

This paper presents a novel approach for the detection and recognition of traffic signs. Color images are acquired by a camera mounted in a car. In the first step these images are color segmented with a pixel classifier. Color combinations which are characteristic for traffic signs generate hypotheses. These hypotheses are verified using a pictogram classifier.

Traffic sign recognition using colour information

In this paper, we present a novel approach for the detection and recognition of traffic signs. Colour images are acquired by a camera mounted in a car. In the first step, these images are colour segmented with a pixel classifier. Colour combinations which are characteristic for traffic signs generate hypotheses. These hypotheses are verified using a pictogram classifier. Our system has been successfully tested on thousands of traffic scenes. The processing of a 512 by 512 frame takes approximately 1 second on a Sparc-10. This project is part of the European research programme PROMETHEUS and is being developed by Daimler-Benz in collaboration with various university labs.

6 – From door to door — principles and applications of computer vision for driver assistant systems

Publisher Summary In this chapter, the achievements in vision-based driver assistance at DaimlerChrysler are described. The chapter presents the systems that have been developed for both highways and urban traffic, and describe principles that have proven robustness and efficiency for image understanding in traffic scenes. The development of computer vision systems for cars is promoted for three main reasons: safety, convenience, and efficiency. At least three guiding principles have emerged for robust vision-based driver assistant systems: (1) vision in cars is vision over time; (2) stereo vision providing 3D information became a central component of robust vision systems; and (3) object recognition can be considered as a classification problem. Besides continuous improvement of the robustness of the image analysis modules, sensor problems have to be overcome. As other information sources, such as radar, digital maps, and communication become available in modern cars, their utilization will help to raise the performance of vision based environment perception.

Reconstruction of topographies form multiple SEM views

Abstract In semiconductor device fabrication the control of individual process steps — such as microlithography and etching — requires the measurement of microstructure topographies by intelligent sensors. The paper outlines a method to reconstruct the three-dimensional surface topography from stereo images, acquired by a scanning electron microscope (SEM).