摘要:
An image processing system and the like capable of accurately recognizing lane edges defined by dotted lane marks are provided. According to an image processing system (100) of the present invention, a first processing unit (110) searches a road surface image captured by a vehicle-mounted camera for a “small area,” which is composed of high- or low-luminance pixels and satisfies “eligibility conditions” on the “size,” “shape,” and “arrangement” in the road surface image. Additionally, a second processing unit (120) recognizes “lane edges” of the lane along which the vehicle travels on the basis of the “small area.” Furthermore, a third processing unit (130) sets a “search range” of the small area searched by the first processing unit (110) on the basis of a result of the foregoing recognition of the lane edges by the second processing unit (120).
摘要:
There is provided an image acquisition means (2) for acquiring a color image of a road via an imaging means (9) mounted on a vehicle (8), a lane mark detection means (3, 4) for performing processing of detecting lane marks of a plurality of predetermined colors different from each other on the road based on color information of the color image, and outputting a result of the processing as lane mark candidate data, and a selection means (6) for selecting lane mark candidate data corresponding to a lane mark defining an actual lane on which the vehicle (8) is traveling from among at least the lane mark candidate data for the respective predetermined colors output from the lane mark detection means (3, 4), and determining and outputting lane data indicating information of the actual lane based on the selected lane mark candidate data. Therefore, even if there are lane marks of different colors on the road, the lane marks of the respective colors can be recognized appropriately from the color image of the road acquired via the imaging means such as a camera.
摘要:
The present invention includes: image capturing means (2) which captures a color image of a road via imaging means (7); area extraction means (3) which extracts areas having a similar color feature value from the captured color image of the road; white balance processing means (5) which performs, for each of the extracted areas, a white balance process of correcting each pixel data in the extracted area so as to reduce the degree of divergence in the level between color components according to a level balance between the color components of the pixel data in the extracted area; and road sign detection means (6) which detects a road sign on the road from the color image subjected to the white balance process. Thus, even if the road illumination state is partially different, it is possible to accurately detect a road sign such as a lane mark from the captured color image of the road via the imaging means such as a camera.
摘要:
An object recognition system mounted on a vehicle comprises a camera and a controller. The camera captures an image in front of the vehicle. The controller divides the image into a plurality of windows. The controller measures distance to the road surface for each of the plurality of windows and thereby obtains a plurality of distances. The controller estimates a relative inclination of the road surface against the vehicle based on the plurality of distances. The controller judges, for each of the plurality of windows, based on the estimated inclination, whether the object is an obstacle or the road surface and thereby obtains a judgment result. The controller recognizes the object based on the judgment result. The system recognizes objects in front of the vehicle even when the vehicle is lopsided by pitching, rolling or else.
摘要:
An object recognition system including a position sensor, an image sensor and a controller is provided. The position sensor determines the position of an object, and the image sensor captures an image of the object. The controller sets a processing area within the image captured by the image sensor based on the position of the object determined by the position sensor and a predetermined size for the object to be recognized. The controller extracts horizontal edges from the processing area, and judges whether each of the extracted edges belongs to the object based on characteristics of the object to be recognized. More practically, the controller can check if each of the extracted edges contains a pixel on a boundary of the processing area. If so, the controller judges to the edge to belong to an object other than the object. The controller can also examine the linearity of each of the extracted edges. If the linearity is larger than a predetermined value, the controller judges the edge to belong to an object other than the object. The controller can also examine the slope of each of the extracted edges. If the slope is larger than a predetermined value, the controller judges the edge to belong to an object other than the object. Thus, the edges judged to belong to an object other than the object are removed. The controller recognizes the outline of the object based on the edges judged to belong to the object.
摘要:
An object recognition system including a position sensor, an image sensor and a controller is provided. The position sensor determines the position of an object, and the image sensor captures an image of the object. The controller sets a processing area within the image captured by the image sensor based on the position of the object determined by the position sensor and a predetermined size for the object to be recognized. The controller extracts horizontal edges from the processing area, and identifies horizontal edges belonging to the outline of the object from the extracted edges. Thus, the object can be recognized based on the horizontal edges only. Similarly, the controller can extract vertical edges from the processing area and identify vertical edges belonging to the outline of the object. Preferably, the controller selects upper, lower, left, and right horizontal and vertical candidate ends of the object from the identified horizontal and vertical edges respectively, from which upper, lower, left, and right ends of the object are determined. If either one or both of the left and right horizontal candidate ends cannot be selected, candidate ends can be estimated based on the position of the object recognized in a previous recognition cycle, and the estimated candidate ends can be used in lieu of the candidate ends selected from the horizontal edges.