Image analysis – Applications – Vehicle or traffic control
Reexamination Certificate
2001-02-15
2004-08-10
Boudreau, Leo (Department: 2621)
Image analysis
Applications
Vehicle or traffic control
C382S106000, C382S154000, C340S901000, C340S903000, C340S937000
Reexamination Certificate
active
06775395
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to an optical object recognition system which detects objects in front of a vehicle such as an automobile, etc., by using an image capturing device having at least two cameras mounted on the vehicle. More specifically, the present invention concerns an object recognition system, which recognizes objects by using a plurality of windows in the captured images.
BACKGROUND OF THE INVENTION
In recent years, devices which determine the distance and size of objects in front of a vehicle, and which appropriately control the vehicle in accordance with this determination, have been proposed for the purpose of improving the safety of vehicle operation.
Laid-open Japanese Patent Application No. Hei 9-79821 describes one example of a system which employs an optical distance measuring device comprising two light-receiving elements to determine whether an object whose distance has been detected is a physical object or a road area (including characters or white lines on the road surface). The system calculates distances for respective calculation areas, and recognizes the areas in which obstructions are present by clustering calculation areas whose mutual distances are within a fixed range and are proximate to each other in the horizontal direction. In the case of this clustering, calculation areas whose distances have not been measured are also clustered.
Japanese unpublished Patent Application No. Hei 11-169567 assigned to the same assignee of the present invention, which was filed on Jun. 16, 1999, describes a system capable of recognizing objects quickly by assigning distance labels to individual windows according to the distance measured for the respective windows and clustering the windows based on the distance labels. The distance labels are predetermined for distance ranges, which in turn are predetermined according to errors in measured distances. Since windows are assigned with distance labels corresponding to the distance ranges to which the distances measured for the respective windows belong, the windows which represent the same object are clustered accurately, allowing the object in front of the vehicle to be recognized more accurately.
Object recognition systems such as the one described above, however, may recognize objects erroneously due to raindrops on the automobile windshield in front of the imaging cameras or due to noise in images. In dealing with this situation, attempts to detect raindrops using technologies for recognizing the outside environment in bad weather by means of visible light cameras, infrared cameras, rain sensors, or other external sensors, as disclosed by Laid-open Japanese Patent Application No. 2000-19259, will incur high costs.
On the other hand, checking the validity of recognizing objects from captured images separately from the object recognition itself in order to prevent raindrops or noise from causing objects to be recognized erroneously will require additional processing time and memory capacity.
Thus, an object of the present invention is to provide an object recognition system which judges the validity of captured images based on the distances measured for windows instead of using separate sensors and which is free from erroneous recognition caused by raindrops or noise.
Another object of the present invention is to provide an object recognition system which can judge the validity of object recognition during the process of object recognition, reducing the requirements for processing time and memory capacity.
SUMMARY OF THE INVENTION
According to one aspect of the invention, an object recognition system having at least two image sensors and a controller that is adapted for measuring the distance from the system to a physical object with respect to respective windows of an image captured by the sensors is provided. The controller is programmed to form clusters by uniting adjacent windows that have similar measured distances, to judge whether each of the clusters is valid or invalid based on the attributes of the cluster. The controller is further programmed to recognize the physical object based on the clusters judged to be valid.
According to one aspect of the invention, the attributes of the cluster include an area of the cluster. The controller is programmed to calculate the area of the cluster based on the number of windows contained in the cluster and measured distance of each of the windows. The controller is programmed to judge that the cluster is valid if the area is larger than a predetermined threshold value.
According to another aspect of the invention, the attributes of the cluster include the number of windows contained in the cluster. The controller is programmed to judge that the cluster is valid if the number of windows contained in the cluster is larger than a threshold value that is predetermined according to the measured distances of the windows contained in the cluster.
According to one aspect of the invention, the controller is further programmed to detect any failed state of the captured image based on the number of clusters judged to be invalid.
According to another aspect of the invention, the controller is further to programmed to detect any failed state of the captured image based on the ratio of the number of clusters judged to be invalid to the total number of clusters contained in the captured image.
The process of recognizing the object or the control of the vehicle mounting the system based on the recognized object may be disabled if said controller judges that the captured image is in the failed state.
According to another aspect of the invention, a method for recognizing a physical object in front of a vehicle is provided. The method comprises capturing an image in front of the vehicle, measuring distance from the vehicle to a physical object with respect to respective windows of the captured image, uniting adjacent windows that have similar measured distances to form clusters, judging whether each of the clusters is valid or invalid based on the attributes of the cluster, and recognizing the physical object based on the clusters judged to be valid.
The attributes of the cluster include an area of the cluster. The step of judging includes calculating the area of the cluster based on the number of windows contained in the cluster and measured distance of each of the windows, and judging that the cluster is valid if the area is larger than a predetermined threshold value.
According to another aspect of the invention, the attributes of the cluster include the number of windows contained in the cluster. The step of judging includes judging that the cluster is valid if the number of windows contained in the cluster is larger than a threshold value that is predetermined according to the measured distances of the windows contained in the cluster.
According to another aspect of the invention, the method for recognizing a physical object further comprises detecting any failed state of the captured image based on the number of clusters judged to be invalid.
According to another aspect of the invention, the method for recognizing a physical object further comprises detecting any failed state of the captured image based on the ratio of the number of clusters judged to be invalid to the total number of clusters contained in the captured image.
The step of recognizing or the control of the vehicle based on the recognized object may be disabled if it is judged that the captured image is in the failed state.
The controller can comprise a micro-controller which typically includes a central unit (CPU), or a micro-processor, a read-only memory (ROM) containing control programs that when executed by the processor performs respective functions which are to be described hereafter. The controller also includes a random-access memory (RAM) that provides a working area for the CPU and temporary storage for various data and programs.
REFERENCES:
patent: 5304980 (1994-04-01), Maekawa
patent: 5589908 (1996-12-01), Irie
patent: 5610993 (1997-03-01), Yamamoto
patent: 5956424 (1999-09-
Aoki Tomoyoshi
Nishigaki Morimichi
Saka Masakazu
Yuhara Hiromitsu
Akhavannik Hussein
Arent & Fox PLLC
Boudreau Leo
Honda Giken Kogyo Kabushiki Kaisha
LandOfFree
Object recognition system does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Object recognition system, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Object recognition system will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3344512