This paper describes the Generic Obstacle and Lane Detection system (GOLD), a stereo vision-based hardware and software architecture to be used on moving vehicles to increment road safety. Based on a full-custom massively parallel hardware, it allows to detect both generic obstacles (without constraints on symmetry or shape) and the lane position in a structured environment (with painted lane markings) at a rate of 10 Hz. Thanks to a geometrical transform supported by a specific hardware module, the perspective effect is removed from both left and right stereo images; the left is used to detect lane markings with a series of morphological filters, while both remapped stereo images are used for the detection of free-space in front of the vehicle. The output of the processing is displayed on both an onboard monitor and a control-panel to give visual feedbacks to the driver. The system was tested on the mobile laboratory (MOBLAB) experimental land vehicle, which was driven for more than 3...

This paper discusses an extension to the Inverse Perspective Mapping geometrical transform to the processing of stereo images and presents the calibration method used on the ARGO autonomous vehicle. The article features also an example of application in the automotive field, in which the stereo Inverse Perspective Mapping helps to speed up the process. 1 Introduction The processing of images is generally performed at different levels, the lowest of which is characterized by the preservation of the data structure after the processing. Different techniques have been introduced for low-level image processing and can be classified in three main categories: Pointwise operations, Cellular Automaton operations, and Global operations [1]. In particular Global operations are transforms between different domains; their application simplifies the detection of image features which, conversely, would require a more complex computation in the original domain. They are not based on a one-to-one map...

This work presents a system for obstacle detection in pair of images acquired by a stereo vision device installed on a moving vehicle. The whole system is structured in a pipeline of two different computational engines: a massively parallel architecture, PAPRICA, devoted to low-level image processing and a traditional serial architecture running medium-level tasks. A geometrical transformation, based on the assumption of a flat road in front of the vehicle, is performed to remove the perspective effect from both images. The difference between the results is used for the detection of freespace in front of the vehicle, thus allowing to avoid the high computational tasks involved in traditional stereo vision approaches; the geometrical transformation is performed by a specific hardware device integrated in PAPRICA architecture. The system was tested on MOB-LAB experimental land vehicle, which was driven for more than 3000 km along extra-urban roads and freeways at speeds up to 80 km/h, a...

This report describes a set of performance metrics and a performance measurement system developed for the evaluation of roadway departure warning systems. The work is part of an effort by the Department of Transportation's National Highway Traffic and Safety Administration (NHTSA) to develop quantitative measures of performance for the evaluation of highway safety systems. The work is also part of an effort to promote dual-use technologies between the Department of Transportation and the Department of Defense. The Department of Commerce's National Institute of Standards and Technology undertook the development of the performance measurement system. The approach for quantitative evaluation rests on establishing a baseline measurement of roadway departure that is of higher accuracy than the warning system under test. The output of the warning system is then compared to the baseline measurement to produce a quantitative measure of performance. The baseline measurement system is installed on a vehicle with the warning system. The baseline system achieves high accuracy by aiming a calibrated camera onto an area of the road adjacent to the vehicle. As the vehicle is driven and roadway departure maneuvers are performed, the output of the video camera, the vehicle's position (from a global positioning system and inertial sensors), and the output of the warning system are synchronized and digitally recorded. After a vehicle run is complete, a suite of software is used to automatically analyze the baseline video data to determine roadway departure. Since the video and other data remain available after the vehicle run, an operator (to establish the credibility of the baseline measurement) can verify a roadway departure. Also, the data can be analyzed to provide insight into the c...

Abstract not found