Temporal coherence in clear path detection
AssigneeGM Global Technology Operations LLC
InventorsWende Zhang, Jinsong Wang
Life of this patent
Estimating the term…
Loading patent…
AssigneeGM Global Technology Operations LLC
InventorsWende Zhang, Jinsong Wang
Estimating the term…
A method of detecting a clear path of travel. Input images are captured at various time step frames. Clear path probability maps of a current and previous time step frames are generated. A corresponding clear path probability map is generated for the current time step frame derived as a function of the clear path probability map of the previous time step frame and of a corresponding mapping that coherently links the previous time step frame to the current time step frame. A weight-matching map is generated. The probability values of the current time step frame are updated as a function of the corresponding probability map. A current frame probability decision map is generated based on updated probability values of the current time step frame. The clear path in the image of the current time step is identified based on the current frame probability decision map.
A method of detecting a clear path of travel, the method comprising the steps of:
Dependencies are inferred from the claim text.
From the US filing history. Foreign family members aren’t listed.
Related to ambient conditions: traffic conditions
Context or environment of the image: exterior to a vehicle by using sensors mounted on the vehicle
Conjoint control of vehicle sub-units of different type or different function
Image and video recognition
About 5 years 6 months of term left · ends 9 May 2032 (estimated)
Drawings are temporarily unavailable. Open the PDF at USPTO
An embodiment relates generally to object detection in a path of travel of a vehicle.
Vision-imaging systems are used in vehicles for enhancing applications within the vehicle such as autonomous driving system or semi-autonomous driving systems. Such systems may be used to autonomously or semi-autonomously control the vehicle through steering systems, throttle control, braking control, or even utilized for lane departure warning systems.
Clear path detection systems identify a clear path in which to operate a vehicle traveling along a road. The path of travel is sensed for objects, including all non-road surfaces, so that the vehicle may travel along an uninterrupted path of travel. Features within an image may be detected and analyzed according to their form and relationship to the roadway. However, such methods may be slow due to intense data processing or the inability to distinguish clear path features and non-clear path features.
FIG. 1 is a vehicle integrating a clear path detection system.
FIGS. 2a, 3a, 4a, 5a illustrate exemplary images captured by an image capture device at various instances of time.
FIGS. 2b, 3b, 4b, 5b illustrate exemplary clear path probability maps at the respective instances of time.
There is shown in FIG. 1 a vehicle 12 integrating a clear path detection system. The clear path detection system includes an image capture device 14 mounted to the vehicle 12. The image capture device 14 is in communication with a processing unit 16.
The image capture device 14 may include a camera or other imaging device that captures images of the road of travel. The image capture device 14 captures images forward of the vehicle that are used to identify a clear path of travel. Preferably, the image capture device 14 is configured for performing continuous image capture. The images captured by the image capture device 14 may be stored in a memory of the image capture device 14, transferred to an off-board memory device, or may be transferred to the processing unit 16 for analysis. Moreover, more than one capture image device may be used in order to capture images rearward and/or on both sides of the vehicle for clear path detection. With a surrounding view system, four wide field-of-view (180 degree +) cameras and a forward looking narrow field-of-view (˜50 degrees) camera perform clear path detection. Images captured rearward of the vehicle may be used in the analysis for identifying the clear path of the road of travel forward of the vehicle.
The captured images are preferably a two-dimensional image of know pixel dimensions. The image contains a plurality of identifiable pixels. Each of the pixels includes a set of bits that correspond to a color on a predetermined color map that represents a color intensity value.