On-vehicle image processor

Patent metadata

Field	Value
Publication number	US-9690996-B2
Application number	US-201314647822-A
Country	US
Kind code	B2
Filing date	Nov 21, 2013
Priority date	Nov 27, 2012
Publication date	Jun 27, 2017
Grant date	Jun 27, 2017

Abstract

An on-vehicle image processor includes an imager mounted on a vehicle, an image convertor and compositor that converts images imaged by the imager into overhead images, a white line candidate area detector that detects a white line candidate area estimated to constitute a parking frame, an endpoint position detector that detects positions of endpoints, a vehicle behavior measurer that measures a movement amount of the vehicle, an endpoint movement position predictor that predicts moved positions of the endpoints detected by endpoint position detector, an endpoint movement amount judger, an endpoint position storage, a distance calculator that judges a degree of proximity of the endpoints, an information output device, an endpoint position oblivion judger that deletes the position of the endpoints satisfying a predetermined condition, a parking frame detector that detects a parking frame from the image, an accelerator opening degree detector and a vehicle behavior controller.

First claim

The invention claimed is:

1. An on-vehicle image processor comprising:
an imager mounted on a vehicle, which monitors circumference of the vehicle and captures images containing at least a road surface;
a white line candidate area detector that detects a white line candidate area which is estimated to form a parking frame from the image imaged by the imager;
an endpoint position detector that detects, from the detected white line candidate area, positions of two endpoints of a center line connecting middle points of lines connecting between each of positive edges in which a brightness of pixels close to a predetermined direction changes to be larger and brighter than a predetermined value and each of negative edges in which a brightness of pixels close to the predetermined direction changes to be larger and darker than a predetermined value, the negative edge being positioned in a predetermined distance from the positive edge;
a vehicle behavior measurer that measures a behavior of the vehicle;
an endpoint movement position predictor that predicts positions in the image to which the two endpoints detected by the endpoint position detector are estimated to be moved during a predetermined time difference, based on a behavior of the vehicle measured by the vehicle behavior measurer;
an endpoint movement amount judger that obtains a difference between the positions of the endpoints predicted by the endpoint movement position predictor and the positions of the endpoints detected by the endpoint position detector and judges whether the difference is larger than a predetermined value;
an endpoint position storage that stores the positions of endpoints when in the endpoint movement amount judger judges that the difference of the positions of the endpoints is larger than the predetermined value;
a parking frame detector that detects a parking frame based on a position of the white line candidate area; and
a distance calculator that calculates a distance between the positions of the endpoints detected by the endpoint position detector or line elements having the endpoints and the positions of the endpoints stored in the endpoint position storage,
wherein, when the distance detected by the distance calculator is smaller than a predetermined value, the parking frame detector reduces a degree of contribution of the white line candidate area containing the endpoints detected by the endpoint position detector when the parking frame detector executes the detection of the parking frame.

2. The on-vehicle image processor according to claim 1 ,
wherein the predetermined time difference is set to be shorter as the vehicle velocity measured by the vehicle behavior measurer is higher.

3. The on-vehicle image processor according to claim 1 ,
wherein, when the vehicle moves in a predetermined distance during the predetermined time difference, the endpoint movement position predictor predicts positions of the endpoints at the time when the vehicle moves in the predetermined distance.

4. The on-vehicle image processor according to claim 1 ,
further comprising an endpoint position oblivion judger that clears the positions of the endpoints stored in the endpoint position storage, when it is judged that the distance calculated in the distance calculator is within a predetermined value over a predetermined time.

5. The on-vehicle image processor according to claim 4 ,
wherein the endpoint position oblivion judger clears the positions of the endpoints stored in the endpoint position storage in a shorter time as a velocity of the vehicle is within a predetermined range and the velocity of the vehicle is in a higher state.

6. The on-vehicle image processor according to claim 4 ,
wherein the endpoint position oblivion judger clears the positions of the endpoints stored in the endpoint position storage in a shorter time as the endpoints stored in the endpoint position storage are closer to a center of the image.

7. The on-vehicle image processor according to claim 4 ,
wherein the endpoint position oblivion judger clears the positions of the endpoints stored in the endpoint position storage in a shorter time as the distance between the endpoints calculated in the distance calculator is closer.

8. The on-vehicle image processor according to claim 4 ,
wherein the endpoint position oblivion judger clears the positions of the endpoints stored in the endpoint position storage in a shorter time as a direction to which the white line candidate area extends is closer to a horizontal direction of the image.

9. The on-vehicle image processor according to claim 1 ,
wherein the predetermined time difference is set to be longer for an area of the image in a lower distance from the imager than a predetermined distance than for an area of the image in a predetermined distance or more from the imager.

10. The on-vehicle image processor according to claim 1 ,
further comprising an adhesive judger that judges a possibility that an adhesive clings to the imager,
wherein, when the possibility is determined to be high, the predetermined time difference is changed.

11. The on-vehicle image processor according to claim 1 ,
further comprising an image convertor that converts the images imaged by the imager into overhead images of the vehicle as viewed from right above,
wherein for the area in the predetermined distance or less from the imager, the detection of the parking frame is executed by use of the converted images, and
for the area in the predetermined distance or more from the imager, the detection of the parking frame is executed by use of the images captured by the imager and not converted by the image convertor.

12. The on-vehicle image processor according to claim 1 ,
wherein, when the distance calculated in the distance calculator is smaller than the predetermined value, the parking frame detector strictly limits a threshold value for detecting the parking frame when detecting the parking frame containing the white line candidate area having the endpoints detected in the endpoint position detector.

Assignees

• Clarion Co Ltd

Inventors

• UTAGAWA Akira
• Kiyohara Masahiro
• Ogata Takehito
• Suzuki Satoshi
• Muramatsu Shoji
• Hayakawa Yasuhisa
• Fukata Osamu
• Morimoto Akira

Classifications

• G06T7/73 Primary

  using feature-based methods · CPC title

• B60R2300/806

  for aiding parking · CPC title

• G06T7/246

  using feature-based methods, e.g. the tracking of corners or segments · CPC title

• G06T2207/30256

  Lane; Road marking · CPC title

• B60R2300/307

  virtually distinguishing relevant parts of a scene from the background of the scene · CPC title