Target recognizing device and target recognizing method

• US 20040066952A1
• Filed: 08/18/2003
• Published: 04/08/2004
• Est. Priority Date: 02/19/2001
• Status: Active Grant

• Current Assignee: Honda Giken Kogyo Kabushiki Kaisha (Honda Motor Company)
• Original Assignee: Honda Giken Kogyo Kabushiki Kaisha (Honda Motor Company)
• Inventors: Hasegawa, Yuji
• Granted Patent: US 7,298,907 B2
• Time in Patent Office: Days
• US Class Current: 382/103
• CPC Class Codes: G06F 18/28 Determining representative reference patterns, e.g. by averaging or distorting; Generating dictionaries
• CPC Class Codes: G06T 7/215 Motion-based segmentation
• CPC Class Codes: G06T 7/269 using gradient-based methods
• CPC Class Codes: G06V 10/772 Determining representative reference patterns, e.g. averaging or distorting patterns; Generating dictionaries

Abstract

A target recognizing device that creates a template concerning a target by itself is provided. The device comprises a robot camera for creating continuous images showing a target and an arithmetic operation unit for controlling the robot camera. A robot camera acquires continuous image information by means of a camera rotatable by a motor. The arithmetic operation unit comprises a template management section for allowing an update part to update a template stored in a template storage part and a target recognition section having an image storage part where the acquired image information is stored, a recognizing part for recognizing a target by using the template, and a motion command part for rotating a camera according to the result of recognition.

Claims

1. A target recognizing device comprising:

a template storing device which stores a template for recognizing a target;

an image acquiring device which acquires continuous images including said target;

a recognition processing device which detects an optical flow between at least two images of said continuous images, to obtain an evaluation function value based on said optical flow; and

an update device which updates said template stored in said template storing device, based on the image including said target acquired by said image acquiring device, until said evaluation function value exceeds a predetermined value.

2. A target recognizing device according to claim 1, further comprising;

a driving device which rotates said image acquiring device, and

an operation instruction device which outputs an operation instruction to said driving device based on the instruction from said recognition processing device,

wherein said operation instruction device outputs an operation instruction for stopping the rotational motion when said evaluation function value exceeds a predetermined value.

3. A target recognizing device according to claim 2 further comprising:

a control signal receiving device which receives a control signal indicating a timing at which said template is updated; and

an identification device which identifies the content of the control signal received by said control signal receiving device and informs said update device of the content,

wherein said update device updates said template corresponding to an identification result of said control signal.

4. A target recognizing device according to claim 3, wherein said recognition processing device obtains said evaluation function value by comparing the degree of approximation of said template and the image including said target, setting a shift quantity with respect to the approximating pixel corresponding to the image including said target, and summing up the shift quantities for all pixels in the whole templates.

5. A target recognizing device according to claim 4, wherein the target recognizing device comprises a group of templates for images acquired while rotating the image acquiring device, and the group of template is updated according to the rotation direction of said image acquiring device.

6. A target recognizing device according to claim 5, wherein said group of templates which are updated while rotating in the same direction comprises at least two types of templates, one of which is added with a positive-polarity weighting value, and the other one of which is added with a negative-polarity weighting value.

7. A target recognizing method comprising:

a template storing step in which a template for recognizing a target is stored;

an image acquiring step in which continuous images including said target are acquired by self-rotational motion;

a recognition processing step in which an optical flow between at least two images of said continuous images is detected in order to obtain an evaluation function value based on said optical flow; and

an update step in which said template stored by said template storing step is updated, based on an image including said target acquired by said image acquiring step, by template update timing instructed from outside, until said evaluation function value exceeds a predetermined value.

8. A target recognition program for executing in a computer comprising the processing steps of:

storing a template in an template storing device for recognizing a target;

acquiring continuous images including said target through a image acquiring device;

detecting an optical flow between at least two images of said continuous images and obtaining an evaluation function value based on said optical flow; and

updating said template stored in said template storing device based on an image including said target acquired by said image acquiring device, until said evaluation function value exceeds a predetermined value.

9. A target recognition program for executing in a computer according to claim 8, said processing step of acquiring continuous images comprising the processing steps of:

outputting an instruction for driving a driving device of an image acquiring device to perform spontaneous rotation motion when an instruction is output from said target recognition device; and

outputting an instruction for stopping the rotational motion of the image acquiring device when said evaluation function value exceeds a predetermined value.

10. A target recognition program for executing in a computer according to claim 9, wherein the program further comprising the processing steps of:

receiving a control signal indicating a timing at which said template is updated, and

identifying the content of the control signal received in said processing step of receiving the control signal and informing the content of the control signal to the update device;

wherein said update device updates said template according to the result of the processing step of identifying n processing of said control signal.

11. A target recognizing program according to claim 10, wherein said processing step of recognizing the target further comprising the steps of:

comparing a degree of approximation of said template and the image including said target;

setting shift quantities for pixels of the approximating target to a pixel of the target in the template; and

summing up shift quantities for all pixels in the whole templates for obtaining said evaluation function value.

12. A target recognition program according to claim 11, wherein a plurality of template groups are provided for use in updating templates according to the rotation directions at the time of image acquiring processing.

13. A target recognition program according to claim 12, wherein said plurality of templates to be updated at the time of rotating in the same direction include at least two templates, one of which is added with a positive-polarity weighting value and the other one of which is added with a negative-polarity weighting value.