Method and system for detecting pedestrians

Patent metadata

Field	Value
Publication number	US-9934428-B2
Application number	US-201414907744-A
Country	US
Kind code	B2
Filing date	Jul 30, 2014
Priority date	Jul 31, 2013
Publication date	Apr 3, 2018
Grant date	Apr 3, 2018

Abstract

The invention relates to a method for detecting a pedestrian ( 27 ) moving in an environmental region of a motor vehicle relatively to the motor vehicle based on a temporal sequence of images ( 18 ) of the environmental region, which are provided by means of a camera of the motor vehicle ( 1 ), wherein characteristic features are extracted from the images ( 18 ) and a plurality of optical flow vectors is determined to the characteristic features of at least two consecutively captured images of the sequence by means of an image processing device of the motor vehicle, which indicate a movement of the respective characteristic features over the sequence, wherein for detecting the pedestrian ( 27 ), several confidence metrics are determined based on the characteristic features and the optical flow vectors, and based on the confidence metrics, it is examined if a preset plausibility check criterion required for the detection of the pedestrian ( 27 ) is satisfied, wherein the pedestrian ( 27 ) is supposed to be detected if the plausibility check criterion with respect to the confidence metrics is satisfied.

First claim

The invention claimed is:

1. A method for detecting a pedestrian moving in an environmental region of a motor vehicle relatively to the motor vehicle based on temporal sequence of images of the environmental region, which are provided by a camera of the motor vehicle, the method comprising:
extracting characteristic features from the images;
determining a plurality of optical flow vectors to the characteristic features of at least two consecutively captured images of the sequence by an image processing device of the motor vehicle,
wherein the plurality of optical flow vectors indicate a movement of the respective characteristic features over the sequence;
determining several confidence metrics for detecting the pedestrian, based on the characteristic features and the optical flow vectors, and based on the confidence metrics, wherein a threshold value is preset for each confidence metric; and
examining whether a preset plausibility check criterion required for the detection of the pedestrian is satisfied, wherein the pedestrian is detected when the plausibility check criterion with respect to the confidence metrics is satisfied, wherein the preset plausibility check criterion includes that all of the confidence metrics pass the respectively associated threshold value.

2. The method according to claim 1 , wherein a plurality of properties of the characteristic features and/or of the optical flow vectors is determined and a confidence metric is respectively defined to each property, the value of which is determined depending on the respective property.

3. The method according to claim 2 , wherein the number of the characteristic features and/or of the optical flow vectors is determined as property.

4. The method according to claim 2 , wherein a magnitude of the optical flow vectors is determined as a property.

5. The method according to claim 2 , wherein a direction of the optical flow vectors is determined as a property.

6. The method according to claim 2 , wherein as a property, the number of characteristic features is determined, the flow vectors of which have the same direction.

7. The method according to claim 2 , wherein the characteristic features are tracked over the sequence of the images, and as a property of the characteristic features, a period of time is determined, which has elapsed since the first detection of the respective feature.

8. The method according to claim 2 , wherein a change of the direction and/or a change of the magnitude of the optical flow vectors over the sequence are determined as a property.

9. The method according to claim 2 , wherein
the number of the flow vectors is determined as a property, which have a magnitude decreasing over the sequence of the images, and/or
the number of the flow vectors is determined as a property, which have a magnitude increasing over the sequence of the images.

10. A camera system for a motor vehicle including a camera for providing a sequence of images of an environmental region of the motor vehicle and including an image processing device configured to perform a method according to claim 1 .

Assignees

• Connaught Electronics Ltd
• Connaught Electronics Ltd

Inventors

• Mcdonald James
• Mcdonald John

Classifications

• G06T7/215 Primary

  Motion-based segmentation · CPC title

• G06T2207/30241

  Trajectory · CPC title

• G06T7/136

  involving thresholding · CPC title

• G06K9/00805

  Physics · mapped topic

• G06K9/00369 Primary

  Physics · mapped topic