Detecting visual information corresponding to an animal

• US 10,152,649 B2
• Filed: 12/01/2016
• Issued: 12/11/2018
• Est. Priority Date: 12/01/2015
• Status: Active Grant

• Current Assignee: MobilEye Vision Technologies Ltd. (Intel Corporation)
• Original Assignee: MobilEye Vision Technologies Ltd. (Intel Corporation)
• Inventors: Shambik, Yakov, Yogev, Noam, Dagan, Erez
• Primary Examiner(s): Huynh, Van D
• Application Number:
  US15/366,570
• Publication Number:
  US 20170154241A1
• Time in Patent Office: 740 Days
• Field of Search: 382103
• CPC Class Codes: B60R 1/00 Optical viewing arrangements; Real-time viewing arrangements for drivers or passengers using optical image capturing systems, e.g. cameras or video systems specially adapted for use in or on vehicles
• CPC Class Codes: B60R 11/04 Mounting of cameras operative during drive; Arrangement of controls thereof relative to the vehicle
• CPC Class Codes: B60R 2011/0026 Windows, e.g. windscreen
• CPC Class Codes: B60R 2300/105 using multiple cameras
• CPC Class Codes: B60R 2300/307 virtually distinguishing relevant parts of a scene from the background of the scene
• CPC Class Codes: B60R 2300/8053 for bad weather conditions or night vision
• CPC Class Codes: B60R 2300/8093 for obstacle warning
• CPC Class Codes: G06V 10/467 Encoded features or binary features, e.g. local binary patterns [LBP]
• CPC Class Codes: G06V 20/58 Recognition of moving objects or obstacles, e.g. vehicles or pedestrians; Recognition of traffic objects, e.g. traffic signs, traffic lights or roads
• CPC Class Codes: G06V 40/10 Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
• CPC Class Codes: G06V 40/103 Static body considered as a whole, e.g. static pedestrian or occupant recognition
• CPC Class Codes: G08G 1/16 Anti-collision systems road vehicle drive control systems for predicting or avoiding probable or impending collision otherwise than by control of a particular sub-unit B60W30/08
• CPC Class Codes: G08G 1/166 for active traffic, e.g. moving vehicles, pedestrians, bikes
• CPC Class Codes: H04N 7/183 for receiving images from a single remote source

Abstract

A system for detecting an animal in a vicinity of a vehicle is disclosed. The system includes at least one processor programmed to receive, from an image capture device, at least one image of the vicinity of the vehicle; analyze the at least one image using a truncated animal appearance template to detect visual information suspected to be associated with an animal, wherein the truncated animal appearance template corresponds to a portion of a body and one or more limbs of a suspected animal; and initiate a vehicle response based on the detected visual information of the suspected animal in the at least one image.

Claims

1. A system for detecting an animal in a vicinity of a vehicle, the system comprising:

at least one processor programmed to;

receive, from an image capture device, at least one first visual image of the vicinity of the vehicle;

analyze the at least one first image using a truncated animal appearance template to detect first visual information suspected to be associated with an animal, wherein the truncated animal appearance template corresponds to a portion of a body and one or more limbs of a suspected animal;

receive, from the image capture device, at least one second visual image of the vicinity of the vehicle, the at least one second image being captured later in time than the at least one first image;

analyze the at least one second image using the truncated animal appearance template to detect second visual information suspected to be associated with the animal; and

initiate a vehicle response based on consistency of the first detected visual information of the suspected animal in the at least one first visual image with the second detected visual information of the suspected animal in the at least one second visual image.

2. The system of claim 1, wherein the at least one processor is further programmed to:

segment the at least one first visual image into a plurality of search areas; and

detect one or more areas of the at least one first visual image that include first visual information that matches the truncated animal appearance template.

3. The system of claim 2, wherein the at least one processor is further programmed to:

determine whether visual information of a prior animal was detected in one or more prior captured images; and

initiate the vehicle response when the first visual information of the suspected animal in the at least one first visual image is consistent with visual information of the prior animal detected in the one or more prior captured images.

4. The system of claim 2, wherein the at least one processor is further programmed to:

generate a data structure corresponding to a detected area of the at least one first visual image that includes first visual information that matches the truncated animal appearance template;

adjust the data structure according to a calculated estimated location of the suspected animal and an estimated change in scale of the suspected animal in a subsequent captured image;

analyze the at least one second visual image with respect to the adjusted data structure to detect areas of the at least one second visual image that include second visual information that matches the adjusted data structure; and

initiate the vehicle response when the at least one second visual image and the adjusted data structure meet a predefined criteria.

5. The system of claim 2, wherein the at least one processor is further programmed to determine a size of at least one of the search areas based on a detected distance of the first visual information in the at least one first visual image from the image capture device.

6. The system of claim 2, wherein the truncated animal appearance template does not correspond to a shape of a head of a suspected animal or a shape of a tail of a suspected animal.

7. The system of claim 2, wherein the at least one processor is further programmed to analyze the at least one first visual image using a plurality of truncated animal appearance templates.

8. The system of claim 2, wherein the truncated animal appearance template corresponds to a suspected animal having proportions characterized by a predetermined torso size and a predetermined limb size.

9. The system of claim 2, wherein the truncated animal appearance template corresponds to one of a plurality of truncated animal appearance templates of the suspected animal at one of a plurality of different viewing angles.

10. The system of claim 2, wherein the truncated animal appearance template corresponds to a suspected animal according to one of a plurality of states of motion including walking, running, and standing still.

11. The system of claim 2, wherein the truncated animal appearance template corresponds to an appearance of a suspected animal among a grouping of two or more animals positioned adjacent to one another.

12. The system of claim 1, wherein the consistency is based on consistency of the first visual information and the second visual information with a kinetic model.

13. The system of claim 1, wherein the at least one processor is further programmed to predict a motion of the suspected animal by applying a kinetic model to the first visual information and the second visual information.

14. The system of claim 13, wherein the vehicle response is based on the predicted motion of the suspected animal.

15. A method of detecting an animal in a vicinity of a vehicle, the method comprising:

receiving a first plurality of visual images of the vicinity of the vehicle from one or more image capture devices;

receiving a truncated animal appearance template, wherein the truncated animal appearance template corresponds to appearance of at least a portion of a body and one or more limbs of an animal without a shape of a head of the animal;

processing the first plurality of visual images using the truncated animal appearance template to detect, in at least one image from the first plurality of images, first visual information that corresponds to the truncated animal appearance template;

receiving, from the one or more image capture devices, a second plurality of visual images of the vicinity of the vehicle, the second plurality of visual images being captured later in time than the first plurality of visual images;

processing the second plurality of visual images using the truncated animal appearance template to detect, in at least one image from the second plurality of images, second visual information that corresponds to the truncated animal appearance template; and

initiating a vehicle response when visual information corresponding to the truncated animal appearance template is detected in the at least one image of the first plurality of images and the at least one image of the second plurality of images.

16. The method according to claim 15, wherein initiating the vehicle response comprises initiating at least one of an alert or a motion control instruction for the vehicle.

17. A system for detecting an animal in a vicinity of a vehicle, comprising:

a camera configured to acquire a first plurality of visual images of the vicinity of the vehicle during a first time period and a second plurality of visual images of the vicinity of the vehicle during a second time period later than the first time period;

a storage unit storing a truncated animal appearance template, wherein the truncated animal appearance template corresponds to an appearance of at least a portion of a body and limbs of an animal without a shape of a head of the animal; and

a processor configured to;

process the first plurality of visual images to detect in at least one of the first plurality of images first visual information that corresponds to the truncated animal appearance template;

process the second plurality of visual images to detect, in at least one of the second plurality of images second visual information that corresponds to the truncated animal appearance template; and

initiate a response when visual information corresponding to the truncated animal appearance template is detected in the at least one image of the first plurality of images and the at least one image of the second plurality of images.

18. The system of claim 17, wherein the processor is further configured to:

analyze a current captured image of the first plurality of images by segmenting the current image into a plurality of search areas; and

use template matching to detect areas of the current image that include first visual information that matches the truncated animal appearance template, wherein a size of the search area is based on a detected distance of the first visual information in the current image from the camera.

19. The system of claim 18, wherein the processor is further configured to:

determine whether visual information of a suspected animal was detected in one or more prior captured images of the plurality of images; and

initiate the response when the first visual information of the current image is determined, according to predefined criteria, to be consistent with visual information of a suspected animal detected in the one or more prior captured images.

20. The system of claim 18, wherein the processor is further configured to:

generate a data structure corresponding to a detected area of a first image of the first plurality of images that includes first visual information that matches a truncated animal appearance template;

adjust the data structure according to a calculated estimated location of the suspected animal and an estimated change in scale of the suspected animal in a subsequent image;

analyze a second image of the second plurality of images with respect to the adjusted data structure to detect areas of the subsequent image that include visual information that matches the adjusted data structure; and

initiate the response when the second image and the adjusted data structure meet a predefined criteria.

21. The system of claim 17, wherein the processor is configured to process the first and the second plurality of images using a plurality of truncated animal appearance templates.

22. The system of claim 17, wherein the truncated animal appearance template corresponds to a suspected animal having proportions characterized by a predetermined torso size and a predetermined limb size.

23. The system of claim 17, wherein the truncated animal appearance template corresponds to a suspected animal according to one of a plurality of states of motion including walking, running, and standing still.