Abstract:
In vehicle-lane boundary line detection, low-luminance values are acquired from areas corresponding to below a tire and directly below a vehicle center based on a road surface image. A snow rut degree is calculated based on a luminance ration between the areas. A probability is calculated from a map based on the calculated snow rut degree. A parameter indicating the degree of snow rut likeness is calculated by a low-pass filtering process. A snow rut determination is made by the calculation result being compared with a predetermined threshold. A final determination of whether or not a snow rut is present is made, with reference to an outside temperature. When determined that a snow rut is present, a determination is made not to perform the detection. When determined that a snow rut is not present, a determination is made to perform the detection.
Abstract:
A cruising zone division line recognition apparatus has an image acquisition device that acquires an image including a road surface ahead of a vehicle, and an image recognition device. The image recognition device adds blurring to an area including the road surface in the acquired image and recognizes a cruising zone division line from the image to which blurring has been added. When blurring is added, a cruising zone division line that is an intermittent double line included in a captured image can be made unclear. Therefore, the recognized cruising zone division line can be prevented from becoming a discontinuous, disjointed line.
Abstract:
The driving support system extracts edge points based on luminance of pixels, in which up edge point is the edge point where the luminance of the inner pixel is smaller than that of the outer pixel, down edge point is the edge point where the luminance of the outer pixel is smaller than the luminance of inner pixel. The system determines, among a plurality of line candidates acquired in accordance with the edge points location, a lane candidate excluding the line candidate that satisfies a predetermined exclusion condition, to be the lane marking, the line candidate determined as the lane marking being located most closely to the vehicle position. The exclusion condition includes a condition where the number of edge points of the up edge line is larger than or equal to a predetermined point threshold, compared to the number of edge points of the down edge line.
Abstract:
A road parameter calculator is provided which is equipped with an edge-point extracting unit, a road parameter calculating unit, a gradient detecting unit, and a modeling unit. The image acquiring unit. The edge-point extracting unit extracts edge points from an image of a frontal view of a vehicle. The parameter calculating unit calculates a road parameter using the edge points through a Kalman filter. The gradient detecting unit detects a change in gradient of the road in front of the vehicle. The modeling unit is responsive to a change in gradient to make a model as extending more straight than when the change in gradient is not detected. This minimizes adverse effects of the change in gradient of the road on the calculation of the road parameter.
Abstract:
A road parameter estimation apparatus estimates road parameters and includes an image acquiring unit, an edge point extracting unit, an area setting unit, an estimating unit, and a vehicle speed acquiring unit. The image acquiring unit acquires an image that shows an area ahead of the vehicle. The edge point extracting unit extracts edge points in the image. The area setting unit sets an area in the image. The area is a part of the image and having a far-side borderline as a boundary thereof The far-side borderline is a virtual line that is ahead of the vehicle by a distance. The estimating unit estimates road parameters using a Kalman filter based on the edge points. The vehicle speed acquiring unit acquires a vehicle speed. The area setting unit increases the distance from the vehicle to the far-side borderline of the area in the image as the vehicle speed increases.
Abstract:
An in-vehicle system as a road curvature detection device calculates a curvature of a road in front of a vehicle based on an acquired front scene image. The in-vehicle system receives gradient information from the data map. The gradient information corresponds to a current road section on the road on which the vehicle drives. The in-vehicle system detects a gradient accuracy of the received gradient information. When the current road section has a gradient, i.e. the road is an uphill or downhill road, the in-vehicle system selects an appropriate special detection methods based on the gradient accuracy of the received gradient information, and calculates a road curvature by using the selected special detection method. Each of the special detection methods calculates a road curvature while effectively suppressing influence of a road gradient indicated by the received gradient information.
Abstract:
This invention is provided with: a camera for capturing the image of a travel path; an edge point extraction unit for extracting edge points on the basis of the brightness of an image captured by the camera; a candidate line extraction unit for extracting, on the basis of the succession of the extracted edge points, a candidate line for a boundary line demarcating the travel path; a frequency calculation unit for calculating, on the basis of edge points belonging to the candidate line extracted by the candidate line extraction unit, the frequency distribution of the edge points for a parameter that specifies the width of the boundary line; a probability generation unit for calculating, on the basis of the frequency distribution calculated by the frequency calculation unit, the distribution for the probability that the candidate line at the parameter is the boundary line; and a boundary line recognition unit for recognizing the boundary line on the basis of the probability distribution calculated by the probability generation unit.
Abstract:
In a travel division line recognition apparatus, an extracting unit extracts a travel division line candidate from an image of a surrounding environment including a road, captured by an on-board camera. A calculating unit calculates a degree of reliability that the extracted travel division line candidate will be the travel division line. A recognizing unit selects the travel division line candidate based on the calculated degree of reliability, and recognizes the travel division line using the selected travel division line candidate. In the calculating unit, a solid object processing unit recognizes solid objects including vehicles, sets a suppression area including a frontal-face suppression area covering a frontal face of the other vehicle and a side-face suppression area covering a side face of the other vehicle, based on the recognized solid objects, and reduces the degree of reliability of the travel division line candidate present within the suppression area.
Abstract:
In an apparatus for recognizing a lane partition line, a pre-branch section setting unit sets a pre-branch section of a road on which an own vehicle is traveling. The pre-branch section extends a section length from a start line that is positioned closer to the own vehicle than a branch location by an offset distance in a traveling direction of the own vehicle. A determination unit determines whether or not a geographical location of the own vehicle is within the pre-branch section. If it is determined that the geographical location of the own vehicle is within the pre-branch section, a recognition unit suppresses recognition of the lane partition line on an diverging road side of the road on which the own vehicle is traveling. The diverging road is another road branching off from the road on which the vehicle is traveling at the branch location.
Abstract:
An apparatus for recognizing a lane line. In the apparatus, when a three-dimensional object lies in the same lane as a subject vehicle and a distance between the three-dimensional object and the subject vehicle is small in an image acquired by an image capture unit, a masking area setter sets a masking area that is partially truncated at or near a lower end of the three-dimensional object in the image. A degree-of-belief calculator is configured to, for each of the edge points extracted by the edge-point extractor, calculate a degree of belief that the edge point is on the lane line. The degree of belief when the edge point is in the masking area is set less than the degree of belief when the edge point is outside the masking area. A lane-line recognizer is configured to recognize the lane line based on the degrees of belief calculated for the edge points.