Information processing apparatus, image capturing apparatus, control system applicable to moveable apparatus, information processing method, and storage medium of program of method

• US 9,794,543 B2
• Filed: 02/22/2016
• Issued: 10/17/2017
• Est. Priority Date: 03/02/2015
• Status: Active Grant

• Current Assignee: Ricoh Company Limited
• Original Assignee: Ricoh Company Limited
• Inventors: Sekiguchi, Hiroyoshi, Yokota, Soichiro, Watanabe, Eita, Kuromizu, Reiko, Takahashi, Yuji
• Primary Examiner(s): TSAI, TSUNG YIN
• Application Number:
  US15/049,971
• Publication Number:
  US 20160261848A1
• Time in Patent Office: 603 Days
• Field of Search: None
• CPC Class Codes: G06F 18/22 Matching criteria, e.g. proximity measures
• CPC Class Codes: G06F 18/24 Classification techniques
• CPC Class Codes: G06V 10/16 using multiple overlapping images; Image stitching
• CPC Class Codes: G06V 10/7515 Shifting the patterns to accommodate for positional errors
• CPC Class Codes: G06V 20/56 exterior to a vehicle by using sensors mounted on the vehicle
• CPC Class Codes: G06V 20/64 Three-dimensional objects
• CPC Class Codes: H04N 13/128 Adjusting depth or disparity
• CPC Class Codes: H04N 13/214 using spectral multiplexing
• CPC Class Codes: H04N 2013/0081 Depth or disparity estimation from stereoscopic image signals

Abstract

An information processing apparatus for processing stereo images of an object includes a disparity calculator to calculate a disparity of the stereo images. The disparity calculator includes an evaluation value calculator to calculate a correlation evaluation value in a search range set for the stereo images; an extreme value detector to detect an extreme value of the correlation evaluation value; a threshold setting unit to set a threshold range for the detected extreme value; a counting unit to count the number of the extreme value existing within the threshold range; and an updating unit to update the set threshold range to a new threshold range when a new extreme value indicating a higher correlation level of the disparity compared to the set threshold range is detected. The counting unit counts the number of the extreme value in a threshold range updated most recently in the search range.

Claims

1. An information processing apparatus for processing stereo images of an object captured from a plurality of viewpoints to recognize the object, the information processing apparatus comprising:

processing circuitry to calculate a disparity of the stereo images by using a matching process, the processing circuitry being configured to;

calculate a correlation evaluation value in a search range set for the stereo images;

detect an extreme value of the calculated correlation evaluation value;

set a threshold range for the detected extreme value;

count a number of the extreme value existing within the threshold range; and

update the set threshold range to a new threshold range when a new extreme value indicating a higher correlation level of the disparity compared to the set threshold range is detected,

wherein the processing circuitry is configured to count the number of the extreme value in the new threshold range updated most recently in the search range.

2. The information processing apparatus of claim 1, wherein the correlation evaluation value indicates a dissimilarity level, and the extreme value is a local minimum of the correlation evaluation value.

3. The information processing apparatus of claim 1, wherein the processing circuitry is further configured to update the set threshold range to the new threshold range by setting the new extreme value having the higher correlation level of the disparity at the center of the new threshold range.

4. The information processing apparatus of claim 1, wherein the processing circuitry is further configured to determine a null of a disparity corresponding to an extreme value existing within the most recently updated threshold range when a counting number becomes a given value or more.

5. The information processing apparatus of claim 1, when a correlation evaluation value indicating a higher correlation level of the disparity compared to the most recently updated threshold range is detected, the processing circuitry is configured to output a disparity corresponding to the correlation evaluation value indicating the higher correlation level of the disparity.

6. The information processing apparatus of claim 5, wherein the processing circuitry is further configured to reset the number.

8. A control system applicable to a moveable apparatus, comprising:

an image capture sensor mountable on the moveable apparatus to capture stereo images of an object from a plurality of viewpoints;

the information processing apparatus of claim 1 to process the stereo images to recognize the object; and

a control circuit to control the moveable apparatus based on a recognition result of the information processing apparatus.

7. An image capturing apparatus, comprising:

an image capturing sensor to capture an image of an object from a plurality of viewpoints to acquire stereo images of the object; and

an information processing apparatus to process the stereo images to recognize the object, the information processing apparatus comprising;

processing circuitry to calculate a disparity of the stereo images by using a matching process, the disparity calculator comprising processing circuitry being configured to;

calculate a correlation evaluation value in a search range set for the stereo images;

detect an extreme value of the calculated correlation evaluation value;

set a threshold range for the detected extreme value;

count a number of the extreme value existing within the threshold range; and

update the set threshold range to a new threshold range when a new extreme value indicating a higher correlation level of the disparity compared to the set threshold range is detected,

wherein the processing circuitry is configured to count the number of the extreme value in the new threshold range most recently updated in the search range.

9. A method of recognizing an object by capturing stereo images of the object from a plurality of viewpoints and calculating a disparity of the stereo images by using a matching process, the method comprising;

calculating a correlation evaluation value in a search range set for the stereo images;

detecting an extreme value of the calculated correlation evaluation value;

setting a threshold range for the detected extreme value;

counting a number of extreme value existing within the set threshold range; and

updating the set threshold range to a new threshold range when an extreme value having a higher correlation level of the disparity compared to the set threshold range is detected at the detecting step; and

counting the number of extreme value in the new threshold range updated most recently in the search range.

10. A non-transitory storage medium storing a program that, when executed by a computer, causes the computer to execute the method of claim 9.