# US 2011/0085026 A1 · Detection method and detection system of moving object

> Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt

- Number: US 2011/0085026 A1
- Document ID: US20110085026A1
- Status: Published application
- Kind: A1
- Application number: 12/860,110
- Filed: 2010-08-20
- Published: 2011-04-14
- Earliest claimed: 2009-10-12
- Inventors: Ming-Hwei PERNG; Yan-Fang Fan
- Web page: https://patentgrep.com/patent/US20110085026A1

## CPC

- G06T 7/285: using a sequence of stereo image pairs
- G06T 7/97: {Determining parameters from multiple pictures (depth or shape recovery from multiple images G06T7/55; stereo camera calibration G06T7/85)}
- G06T 2207/10021: Stereoscopic video; Stereoscopic image sequence
- G06T 2207/30252: Vehicle exterior; Vicinity of vehicle
- G06T: IMAGE DATA PROCESSING OR GENERATION, IN GENERAL

## Abstract

A detection method of a moving object is provided. The method includes: using a left lens system to capture a first and a second left image, and using a right lens system to capture a first right image and a second right image; subdividing the first right image into plural color blocks; selecting N control points on the first left image, and searching M first corresponding points in the first right image; calculating first depth information according to the control points and the first corresponding point, and calculating a possible area in the second right image to search P second corresponding points; using the first corresponding points and the second corresponding points to calculate a two-dimensional planar conversion parameter, and converting pixels in the color block to new positions to obtain a converted image; and identifying a difference area between the second right image and the converted image as an area.

## Claims

### Claim 1

A detection method of a moving object, applied to a detection system, wherein the detection system comprises a platform, a stereo camera including a left lens system and a right lens system disposed on the platform, and a processing module, the detection method comprises:
(a) using the left lens system to capture a first left image and a second left image at a first time and a second time respectively, and using the right lens system to capture a first right image and a second right image at the first time and the second time respectively;
(b) subdividing the first right image into a plurality of color blocks by the processing module;
(c) selecting N control points on the first left image, and searching M first corresponding points of the N control points in the first right image, wherein N and M are a positive integer respectively, and M is not greater than N by the processing module;
(d) calculating depth information according to each of the control points and the corresponding first corresponding point, and calculating a possible area in the second right image in which the M first corresponding points appear according to the depth information by the processing module;
(e) searching P second corresponding points of the M first corresponding points in the possible area, wherein P is a positive integer not greater than M by the processing module;
(f) using the first corresponding points and the second corresponding points contained in each color block to calculate a two-dimensional planar conversion parameter of a deformation of the color block generated from the first time to the second time, and using the two-dimensional planar conversion parameter to convert all pixels in the color block to new positions to obtain a converted image by the processing module; and
(g) identifying a difference area between the second right image and the converted image as an area in which the moving object is located by the processing module.

### Claim 2 (depends on 1)

The detection method according to claim 1, wherein in (c), the N control points are selected at a fixed interval.

### Claim 3 (depends on 1)

The detection method according to claim 1, wherein the depth information is a distance of the control point and the corresponding first corresponding point relative to the platform.

### Claim 4 (depends on 1)

The detection method according to claim 1, wherein in (d), the possible area is calculated according to the depth information in combination with a maximum speed of ego-motion of the platform.

### Claim 5 (depends on 1)

The detection method according to claim 1, wherein the two-dimensional planar conversion parameter is used to calibrate the plurality of color blocks from the first time to the second time to perform background compensation.

### Claim 6 (depends on 1)

The detection method according to claim 1, wherein in (g), an absolute value of a difference between the second right image and the converted image is obtained through a transient difference method.

### Claim 7 (depends on 6)

The detection method according to claim 6, wherein (g) further comprises:
(g1) screening the difference area according to a gray level threshold to obtain a monochrome image.

### Claim 8

A detection system, for detecting a position of a moving object, comprising:
a platform;
a stereo camera including a left lens system, disposed on the platform, for capturing a first left image and a second left image at a first time and a second time respectively, and a right lens system, disposed on the platform, for capturing a first right image and a second right image at the first time and the second time respectively; and
a processing module, connected to the left lens system and the right lens system, for receiving the first left image, the second left image, the first right image, and the second right image, subdividing the first right image into a plurality of color blocks, selecting N control points on the first left image, and searching M first corresponding points of the N control points in the first right image, wherein N and M are a positive integer respectively, and M is not greater than N; for calculating depth information according to each of the control points and the corresponding first corresponding point, and calculating a possible area in the second right image in which the M first corresponding points appear according to the depth information; for searching P second corresponding points of the M first corresponding points in the possible area, wherein P is a positive integer not greater than M; for using the first corresponding points and the corresponding second corresponding points contained in each color block to calculate a two-dimensional planar conversion parameter of a deformation of the color block generated from the first time to the second time, and using the two-dimensional planar conversion parameter to convert all pixels in the color block to new positions to obtain a converted image; and for identifying a difference area between the second right image and the converted image as an area in which the moving object is located.

### Claim 9 (depends on 8)

The detection system according to claim 8, wherein the processing module selects the N control points at a fixed interval.

### Claim 10 (depends on 8)

The detection system according to claim 8, wherein the depth information is a distance of the control point and the corresponding first corresponding point relative to the platform.

### Claim 11 (depends on 8)

The detection system according to claim 8, wherein the processing module calculates the possible area according to the depth information in combination with a maximum speed of ego-motion of the platform.

### Claim 12 (depends on 8)

The detection system according to claim 8, wherein the two-dimensional planar conversion parameter is used to calibrate the plurality of color blocks from the first time to the second time to perform background compensation.

### Claim 13 (depends on 8)

The detection system according to claim 8, wherein the processing module obtains an absolute value of a difference between the second right image and the converted image through a transient difference method.

### Claim 14 (depends on 13)

The detection system according to claim 13, wherein the processing module further screens the difference area according to a gray level threshold to obtain a monochrome image.
