19-03-2014, 04:40 PM
Lane detection algorithm for an onboard camera
[attachment=60273]
ABSTRACT
After analysing the major causes of injuries and death on roads, it is understandable that one of the main goals in the
automotive industry is to increase vehicle safety. The European project SPARC (Secure Propulsion using Advanced
Redundant Control) is developing the next generation of trucks that will fulfil these aims.
The main technologies that will be used in the SPARC project to achieve the desiderated level of safety will be
presented. In order to avoid accidents in critical situations, it is necessary to have a representation of the environment of
the vehicle. Thus, several solutions using different sensors will be described and analysed.
Particularly, a division of this project aims to integrate cameras in automotive vehicles to increase security and prevent
driver's mistakes. Indeed, with this vision platform it would be possible to extract the position of the lane with respect to
the vehicle, and thus, help the driver to follow the optimal trajectory. A definition of lane is proposed, and a lane
detection algorithm is presented. In order to improve the detection, several criteria are explained and detailed.
Regrettably, such an embedded camera is subject to the vibration of the truck, and the resulting sequence of images is
difficult to analyse. Thus, we present different solutions to stabilize the images and particularly a new approach
developed by the "Laboratoire de Production Microtechnique". Indeed, it was demonstrated in previous works that the
presence of noise can be used, through a phenomenon called Stochastic Resonance. Thus, instead of decreasing the
influence of noise in industrial applications, which has non negligible costs, it is perhaps interesting to use this
phenomenon to reveal some useful information, such as for example the contour of the objects and lanes.
SPARC technologies
There are two main elements in the SPARC project, the first is the X-by-wire technology that enables the steering wheel
of the driver to be mechanically disconnected from the wheels. Thus, automotive industry suppresses the steering
column and replaces it by wires, as in plane and aircraft technology. This equipment has been successfully tested in the
Powertrain Equipped with Intelligent Technologies project (PEIT, contract no.: IST-2000-28722), and allows some
automatic controls to maintain drive stability and reduce braking distance in critical situations.
State of the art for lane detection and object recognition
In recent development, one of the most frequently implemented solutions for road recognition is the use of camera based
systems. Indeed, with the actual development of camera sensors, the cost of such devices is decreasing; furthermore, the
information content of vision sensors is extremely rich. In addition, such devices have several other advantages such as,
small sweep time and are passive sensors which avoid interferences with other sensors and users in the environment.
Thus, the GOLD1, ARGO4, and RALPH5 projects were all based on such principles, using one or two cameras
depending on the chosen approach. However, it is clear that such systems are mainly affected by shadows, bad weather
such as fog, rain, and by changing illumination. Indeed, the contrast of the objects contained in the camera scene is
strongly related to illumination condition and a variation of this value can result in the impossibility to detect the desired
objects.
LANE DETECTION
The main aims of the camera are to detect the position of the truck with respect to the vehicle's lane, and perform the
detection and tracking of obstacles. In order to carry out an analysis of the problem of lane detection, it is absolutely
necessary to have a clear understanding of what has to be detected. Thus, the most difficult part of the algorithm is to
define the hypotheses to detect the correct lane. To decrease time consumption of the algorithm, we will define that this
Lane Detection Algorithm‡ will work on specific and limited region of interest§ (12 reduced ROIs are drawn in green in
Figure 4). Thus, all computation algorithms are bounded to small regions and not to the whole image.
CONCLUSIONS
Vehicle safety is one of the most active area of research for the automotive industries. Therefore, solutions for
"intelligent" vehicles are pointing out. From this approach, the European FP6 "Secure Propulsion using Advanced
Redundant Control" (SPARC, EU contract no.: 507859) project is integrating such technologies to achieve safety needs.
Thus, the vehicle wills integer X-by-wire technologies, which allow an autonomous co-pilot to assist the driver in
critical situations.
In order to build a reliable knowledge of the environment of the vehicle, the co-pilot will fuse the information coming
from several sensors. This paper describes how fusion can help to achieve the desired aims of safety of SPARC project.
A specific task within the project is the extraction of pertinent information from an on-board camera. One of its main
objectives is to detect the driver's lane in order to determine if driver's trajectory is the safest. Thus, a definition of
"lane" has been provided, in order to determine precisely what the algorithm of lane detection is expected to find. In
order to see the evolving scene, the camera and the processing platform have to be capable of taking and processing
rapidly the sequence of images. A solution has been proposed to avoid large computation on the whole image, and
different propositions of lane detection have been described. Several criteria had also been developed to weight the
reliability of the lane detection algorithm. Finally, different solutions to reduce the influence of vibrations due to the
truck movement have been proposed and will be tested in the months ahead.
[attachment=60273]
ABSTRACT
After analysing the major causes of injuries and death on roads, it is understandable that one of the main goals in the
automotive industry is to increase vehicle safety. The European project SPARC (Secure Propulsion using Advanced
Redundant Control) is developing the next generation of trucks that will fulfil these aims.
The main technologies that will be used in the SPARC project to achieve the desiderated level of safety will be
presented. In order to avoid accidents in critical situations, it is necessary to have a representation of the environment of
the vehicle. Thus, several solutions using different sensors will be described and analysed.
Particularly, a division of this project aims to integrate cameras in automotive vehicles to increase security and prevent
driver's mistakes. Indeed, with this vision platform it would be possible to extract the position of the lane with respect to
the vehicle, and thus, help the driver to follow the optimal trajectory. A definition of lane is proposed, and a lane
detection algorithm is presented. In order to improve the detection, several criteria are explained and detailed.
Regrettably, such an embedded camera is subject to the vibration of the truck, and the resulting sequence of images is
difficult to analyse. Thus, we present different solutions to stabilize the images and particularly a new approach
developed by the "Laboratoire de Production Microtechnique". Indeed, it was demonstrated in previous works that the
presence of noise can be used, through a phenomenon called Stochastic Resonance. Thus, instead of decreasing the
influence of noise in industrial applications, which has non negligible costs, it is perhaps interesting to use this
phenomenon to reveal some useful information, such as for example the contour of the objects and lanes.
SPARC technologies
There are two main elements in the SPARC project, the first is the X-by-wire technology that enables the steering wheel
of the driver to be mechanically disconnected from the wheels. Thus, automotive industry suppresses the steering
column and replaces it by wires, as in plane and aircraft technology. This equipment has been successfully tested in the
Powertrain Equipped with Intelligent Technologies project (PEIT, contract no.: IST-2000-28722), and allows some
automatic controls to maintain drive stability and reduce braking distance in critical situations.
State of the art for lane detection and object recognition
In recent development, one of the most frequently implemented solutions for road recognition is the use of camera based
systems. Indeed, with the actual development of camera sensors, the cost of such devices is decreasing; furthermore, the
information content of vision sensors is extremely rich. In addition, such devices have several other advantages such as,
small sweep time and are passive sensors which avoid interferences with other sensors and users in the environment.
Thus, the GOLD1, ARGO4, and RALPH5 projects were all based on such principles, using one or two cameras
depending on the chosen approach. However, it is clear that such systems are mainly affected by shadows, bad weather
such as fog, rain, and by changing illumination. Indeed, the contrast of the objects contained in the camera scene is
strongly related to illumination condition and a variation of this value can result in the impossibility to detect the desired
objects.
LANE DETECTION
The main aims of the camera are to detect the position of the truck with respect to the vehicle's lane, and perform the
detection and tracking of obstacles. In order to carry out an analysis of the problem of lane detection, it is absolutely
necessary to have a clear understanding of what has to be detected. Thus, the most difficult part of the algorithm is to
define the hypotheses to detect the correct lane. To decrease time consumption of the algorithm, we will define that this
Lane Detection Algorithm‡ will work on specific and limited region of interest§ (12 reduced ROIs are drawn in green in
Figure 4). Thus, all computation algorithms are bounded to small regions and not to the whole image.
CONCLUSIONS
Vehicle safety is one of the most active area of research for the automotive industries. Therefore, solutions for
"intelligent" vehicles are pointing out. From this approach, the European FP6 "Secure Propulsion using Advanced
Redundant Control" (SPARC, EU contract no.: 507859) project is integrating such technologies to achieve safety needs.
Thus, the vehicle wills integer X-by-wire technologies, which allow an autonomous co-pilot to assist the driver in
critical situations.
In order to build a reliable knowledge of the environment of the vehicle, the co-pilot will fuse the information coming
from several sensors. This paper describes how fusion can help to achieve the desired aims of safety of SPARC project.
A specific task within the project is the extraction of pertinent information from an on-board camera. One of its main
objectives is to detect the driver's lane in order to determine if driver's trajectory is the safest. Thus, a definition of
"lane" has been provided, in order to determine precisely what the algorithm of lane detection is expected to find. In
order to see the evolving scene, the camera and the processing platform have to be capable of taking and processing
rapidly the sequence of images. A solution has been proposed to avoid large computation on the whole image, and
different propositions of lane detection have been described. Several criteria had also been developed to weight the
reliability of the lane detection algorithm. Finally, different solutions to reduce the influence of vibrations due to the
truck movement have been proposed and will be tested in the months ahead.