Automatic vehicle leak detection

Patent metadata

Field	Value
Publication number	US-10372996-B2
Application number	US-201615380930-A
Country	US
Kind code	B2
Filing date	Dec 15, 2016
Priority date	Dec 15, 2016
Publication date	Aug 6, 2019
Grant date	Aug 6, 2019

Abstract

A vehicle controller receives images from a camera upon arrival and upon departure. A location of the vehicle may be tracked and images captured by the camera may be tagged with a location. A departure image may be compared to an arrival image captured closest to the same location as the arrival image. A residual image based on a difference between the arrival and departure images is evaluated for anomalies. Attributes of the anomaly such as texture, color, and the like are determined and the anomaly is classified based on the attributes. If the classification indicates an automotive fluid, then an alert is generated.

First claim

The invention claimed is:

1. A method comprising, by a vehicle controller:
receiving a stream of images from a forward facing camera, each image of the stream of images being received at a vehicle location;
storing the image stream as stored images such that each image of the stream of images is stored as a stored image tagged with the vehicle location at which the each image was received;
detecting parking of the vehicle;
in response to detecting parking of the vehicle:
detecting a parked location of the vehicle; and
identifying an arrival image from the stored images from the stream of images as being received prior to detecting parking of the vehicle and having the vehicle location with which the arrival image is tagged at a predetermined distance from the parked location;

detecting departure of the vehicle from the parked location;
in response to detecting departure of the vehicle from the parked location:
identifying a departure image in the image stream received after departure and having the vehicle location at which the each image was received within a threshold distance from the vehicle location with which the arrival image is tagged;

identifying an anomaly according to an evaluation of the arrival image and the departure image by evaluating a representation of a surface located under the vehicle while parked and that is visible in the arrival image and the departure image; and
determining that the anomaly indicates a fluid leak.

2. The method of claim 1 , wherein identifying the anomaly according to the evaluation of the arrival image and the departure image comprises:
generating a residual image according to a difference between the arrival image and the departure image; and
identifying the anomaly according to analysis of the residual image.

3. The method of claim 2 , wherein identifying the anomaly according to the evaluation of the arrival image and the departure image comprises:
performing image stabilization on a first output of the camera to obtain the arrival image; and
performing image stabilization on a second output of the camera to obtain the departure image.

4. The method of claim 1 , wherein determining that the anomaly indicates the fluid leak comprises performing texture analysis of the anomaly.

5. The method of claim 4 , wherein determining that the anomaly indicates the fluid leak comprises classifying a result of the texture analysis of the anomaly.

6. The method of claim 1 , wherein identifying the anomaly according to the evaluation of the arrival image and the departure image comprises evaluating the arrival image and departure image by a machine learning algorithm.

7. The method of claim 1 further comprising determining the vehicle location for each image of the stream of image by receiving a global positioning (GPS) coordinate from a GPS receiver and movement information from an inertial measurement unit (IMU).

8. The method of claim 1 , further comprising generating an alert in response to determining that the anomaly indicates a fluid leak.

9. A vehicle comprising:
a camera, the camera being forward facing;
a vehicle component containing a fluid;
a controller operably coupled to the camera, the controller programmed to—
receive an image stream from the camera, each image of the image stream being received at a vehicle location;
storing the image stream as stored images such that each image of the image stream is stored as a stored image tagged with the vehicle location at which the each image was received;
detect parking of the vehicle;
in response to detecting parking of the vehicle at a parking location:
identify a set of arrival images from the stored images that were received before detecting parking of the vehicle and that are tagged with vehicle locations in a region about a predetermined distance from the parking location;

detect departure of the vehicle from the parking location;
in response to detecting departure of the vehicle from the parking location, identify a set of departure images from the image stream that were received after detecting departure of the vehicle and received at vehicle locations corresponding to the vehicle locations with which the set of arrival images are tagged;
stabilize the set of arrival images to obtain a stabilized arrival image;
stabilize the set of departure images to obtain a stabilized departure image
evaluate whether (a) the stabilized arrival image and the stabilized departure image indicate an anomaly at the parking location on a surface located under the vehicle while parked;
evaluate whether (b) the anomaly indicates a fluid leak;
if (b), generate an alert.

10. The vehicle of claim 9 , wherein the controller is programmed to evaluate (a) by:
generating a residual image according to a difference between the stabilized arrival image and the stabilized departure image; and
evaluate (c) whether the residual image indicates the anomaly.

11. The vehicle of claim 9 , wherein the controller is further programmed to evaluate (b) by performing texture analysis of the anomaly.

12. The vehicle of claim 11 , wherein the controller is further programmed to evaluate (b) by classifying a result of the texture analysis of the anomaly.

13. The vehicle of claim 9 , wherein the controller is further programmed to evaluate (a) by evaluating the stabilized arrival image and the stabilized departure image by a machine learning algorithm.

14. The vehicle of claim 9 , further comprising a global positioning system (GPS) receiver and an inertial measurement unit (IMU), the controller programmed to determine the vehicle location for each image in the image stream from outputs of the GPS receiver and the IMU.

15. The vehicle of claim 9 , wherein the vehicle component is at least one of an engine, transmission radiator, power braking system, and power steering system.

Assignees

• Ford Global Tech Llc

Inventors

• Jain Jinesh J
• Banvait Harpreetsingh
• Jales Costa Bruno Sielly
• Micks Ashley Elizabeth

Classifications

• G06K9/00771 Primary

  Physics · mapped topic

• G06T2207/10004

  Still image; Photographic image · CPC title

• G06T2207/30252

  Vehicle exterior; Vicinity of vehicle · CPC title

• G06T7/40

  Analysis of texture (depth or shape recovery from texture G06T7/529) · CPC title

• G06T7/001 Primary

  using an image reference approach · CPC title