IP Library › Granted Patent US 11,282,224
Granted Patent B2
US 11,282,224 · App. 16/344,155 · Granted Mar 22, 2022

Information processing apparatus and information processing method

Inventor: Masaki Uchida (Chiba, JP)
Assignee: Sony Interactive Entertainment Inc.
G06T7/70G06T7/13G06T7/66G06T2207/10012
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 11,282,224
App. No.
16/344,155
Filed
Apr 23, 2019
Granted
Mar 22, 2022
Kind
B2
Art Unit
2661
USPC
382/103
Abstract

A target object detection section detects the region of an image of a target object from a frame image captured as a moving image. A contour search section samples the values of pixels in four diagonal directions in a pixel array, using as a start point a pixel of which the corresponding element targets G (green) for observation in the region. The contour search section then detects coordinate point candidates on a contour line of the target object on the basis of changes in luminance Y of the sampled pixels. The contour search section filters the coordinate point candidates based on luminance changes in the color component of the target object. The contour search section causes the distribution of the coordinate points to approximate the contour line. A position information generation section acquires position information regarding the target object in the real space using the gravity center and size of the target object image obtained from the contour line.

Claims (28)

1. An information processing apparatus comprising:

a captured image acquisition section configured to acquire data of a captured image of a target object in an extracted image region;

a contour search section configured to obtain a contour line of the image of the target object by searching the captured image for a coordinate point of which a change in luminance relative to a position inside the captured image of the target object in the extracted image region meets a predetermined condition; and

a position information generation section configured to determine the position of the target object in a real space on the basis of the contour line, wherein

the contour search section searches for the coordinate point along a search route in a diagonal direction in an array of pixels inside the captured image of the target object in the extracted image region, using a pixel inside the captured image of the target object in the extracted image region as a start point.

2. The information processing apparatus according to claim 1 , wherein

the captured image acquisition section acquires data of a color image formed by color-interpolating luminance data obtained using a Bayer array on an image plane; and

the contour search section uses a plurality of pixels having the same color in the initial Bayer array as search start points.

3. The information processing apparatus according to claim 2 , wherein the contour search section uses pixels having the color of green in the initial Bayer array as search start points.

4. The information processing apparatus according to claim 1 , wherein the contour search section performs searches in a plurality of diagonal directions to obtain the contour line by causing the distribution of the detected coordinate points to approximate a predictable shape of the target object image using a least-square method.

5. The information processing apparatus according to claim 1 , wherein the contour search section detects the coordinate point at least on the basis of a change in luminance Y relative to the position of the coordinate point in a YUV space representing the colors of the pixels on a search route.

6. The information processing apparatus according to claim 5 , wherein the contour search section filters the coordinate point on the basis of a change in the luminance of a color component of the target object from among the colors of the pixels on the search route.

7. The information processing apparatus according to claim 1 , wherein

the captured image acquisition section acquires data of stereoscopic images captured of the target object from right and left viewpoints; and

the position information generation section calculates a gravity center of each of the stereoscopic images of the target object on the basis of the contour line, and generates the position information based on a parallax between the gravity centers of the stereoscopic images.

8. The information processing apparatus according to claim 1 , wherein the position information generation section calculates a size of the target object image based on the contour line, and generates the position information by comparing the calculated size with an actual size of the target object.

9. The information processing apparatus according to claim 2 , wherein the contour search section translates the start point from one pixel to another to obtain the contour line for each start point in the captured image of the target object at a known distance from an imaging apparatus, and identifies the pixel to be used as the start point of the search on the basis of an error in the resulting position information.

10. An information processing method comprising:

acquiring data of a captured image of a target object in an extracted image region so as to store the data into a memory;

obtaining a contour line of the captured image of the target object by searching the captured image read from the memory for a coordinate point of which a change in luminance relative to a position inside the captured image of the target object in the extracted image region meets a predetermined condition; and

determining the position of a target object in a real space on the basis of the contour line, wherein

the obtaining the contour line searches for the coordinate point along a search route in a diagonal direction in an array of pixels inside the captured image of the target object in the extracted image region, using a pixel inside the captured image of the target object in the extracted image region as a start point.

11. A non-transitory, computer-readable storage medium containing a computer program, which when executed by a computer, causes the computer to carry out actions, comprising:

acquiring data of a captured image of a target object in an extracted image region;

obtaining a contour line of the captured image of the target object by searching the captured image read from the memory for a coordinate point of which a change in luminance relative to a position inside the captured image of the target object in the extracted image region meets a predetermined condition; and

determining the position of a target object in a real space on the basis of the contour line, wherein

the obtaining the contour line searches for the coordinate point along a search route in a diagonal direction in an array of pixels inside the captured image of the target object in the extracted image region, using a pixel inside the captured image of the target object in the extracted image region as a start point.

12. The information processing apparatus according to claim 4 , wherein one of the plurality of directions comprises a path of pixels having a pattern of different colors and another direction comprises a path of pixels having a single color.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Apr 23, 2019
From: UCHIDA, MASAKI
To: SONY INTERACTIVE ENTERTAINMENT INC.
Reel/Frame 048970/0236 →
Priority Claims (1)
JP JP2016-238748 · Dec 8, 2016 · national
Continuity (1)
Related Publication 20200265601A1 · Aug 20, 2020