Rean Isabella Fedriga

Obstacle Detection with Stereo Vision for Off-Road Vehicle Navigation

In this paper we present an artificial vision algorithm for real-time obstacle detection in unstructured environments. The images have been taken using a stereoscopical vision system. The system uses a new approach, of low computational load, to calculate a V-disparity image between left and right corresponding images, in order to estimate the cameras pitch oscillation caused by the vehicle movement. Then, the obstacles are localized by stereo matching and mapped in real world coordinates. Experimental results on sequences taken from a moving vehicle (which partecipated to the DARPA Grand Challenge 2004) in different unstructured scenarios are then presented, to demonstrate the validity of the approach.

VIAC: An out of ordinary experiment

This paper presents the preliminary results of VIAC, the VisLab Intercontinental Autonomous Challenge, a test of autonomous driving along an unknown route from Italy to China. It took 3 months to run the entire test; all data have been logged, including all data generated by the sensors, vehicle data, and GPS info. This huge amount of information has been packed during the trip, compressed, and transferred back to Parma for further processing. This data is now ready for a deep analysis of the various systems performance, with the aim of virtually running the whole trip multiple times with improved versions of the software. This paper discusses some preliminary figures obtained by the analysis of the data collected during the test. More information will be generated by a deeper analysis, which will take additional time, being the data about 40 terabyte in size.

Pedestrian Detection in Far Infrared Images based on the use of Probabilistic Templates

This article presents a validator stage for a pedestrian detection system based on the use of probabilistic models for the infrared domain. Four different models are employed in order to recognize the pose of the pedestrians; open, almost open, almost closed and fully closed legs are detected. In an attempt to overcome the drawbacks of template-matching in far infrared images, two different approaches are proposed. The algorithm has been tested on an experimental vehicle in different situations and a receiver operating characteristic has been computed.

Pedestrian Detection on a Moving Vehicle: an Investigation about Near Infra-Red Images

The aim of this paper is to investigate the employment of a NIR (near infra-red) camera on a moving vehicle for pedestrian detection, considering the NIR radiation as a suitable waveband for a single sensor device. In particular, we studied a system with special attention to cost, in order to make it affordable for all vehicles. We used as equipment an analog NIR CCD camera and examined the system behaviour in three different lighting conditions: daytime, nighttime using dipped-beam headlights and nighttime using full-beam headlights. Subsequently we chose the lighting situation which proved, in the long run, to be the most suitable for pedestrian detection. The algorithm proved to be efficient in common exurban scenarios

A 13,000 km Intercontinental Trip with Driverless Vehicles: The VIAC Experiment

In recent years the interest in autonomous vehicles has incrementally increased. After the DARPA Challenges new fields of application as agricultural, construction, mining, and also nautical are continuously opening up. In this paper a huge test is presented, the first of this kind in the history of vehicular robotics. A trip from Italy to China with four electric autonomous vehicles will be described focusing on all aspects of the challenge, from the managing issues to the most technical ones. A vehicle-following application (or virtual towing) is the system under test for a three consecutive months and 13,000 km long unique experience.

Sensing requirements for a 13,000 km intercontinental autonomous drive

This paper presents the design issues that were considered for the equipment of 4 identical autonomous vehicles that will drive themselves without human intervention on an intercontinental route for more than 13,000 km.

Test-bed for Unified Perception & Decision Architecture

This paper presents the test-bed that will be developed for a Unified Perception & Decision Architecture (UPDA). Due to the increasing demand of ADAS systems to be mounted on cars, it is more and more important to develop a unified architecture that can communicate and share information between these systems. This is the aim of an ERC-founded project and to develop and test such architecture a car has been set up with many different sensors.

Vision-based monitoring of pedestrian crossings

This paper presents a vision-based intelligent pedestrian crossing able to derive information on the occupancy status of its zebra crossing. The detection of pedestrians can help enhancing safety for road users who need longer to cross the road. At the same time, it produces savings in traffic delay. A stereo vision approach is adopted: the difference between two birds eye views of the crossing is analyzed to extract objects rising up from the road plane. Background subtraction is applied to get rid of noise due to small drifts in the calibration and lens distortion. The discrimination between pedestrians and vehicles is performed taking into account their direction of movement.

Pedestrian Detection in Poor Visibility Conditions: Would SWIR Help?

The 2WIDE_SENSE (WIDE spectral band & WIDE dynamics multifunctional imaging SENSor Enabling safer car transportation) EU funded project is aimed at the development of a low-cost camera sensor for Advanced Driver Assistance Systems (ADAS) applications able to acquire the full visible to Short Wave InfraRed (SWIR) spectrum from 400 to 1700 nm. This paper presents the first results obtained by investigating the SWIR contribution to pedestrian detection in difficult visibility conditions as haze and fog employing the wide-bandwidth camera developed within the project.

Vision technologies for intelligent vehicles

This paper surveys the different technologies involved in the development of computer vision applications for Intelligent Vehicles derived from the 15 year experience of VisLab. Some illustrative examples are also discussed.