Unlock instant, AI-driven research and patent intelligence for your innovation.

Exercisable region detection method and device based on RGBD (Red, Green and Blue Depth)

A technology of area detection and driving area, which is applied in the field of RGBD-based practicable area detection and devices, and can solve problems such as large amount of calculation, difficulty in identifying timeliness, and difficulty in covering all

Inactive Publication Date: 2018-05-08
元橡科技(北京)有限公司
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] There are many problems in this method. For example, road images are ever-changing, and it is difficult to cover all road images through prior learning.
And, to identify the graphic elements in the logical sense of the image, the amount of calculation is very large, and it is difficult to achieve the timeliness of identification

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Exercisable region detection method and device based on RGBD (Red, Green and Blue Depth)
  • Exercisable region detection method and device based on RGBD (Red, Green and Blue Depth)
  • Exercisable region detection method and device based on RGBD (Red, Green and Blue Depth)

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0036] figure 1 It is a flow chart of an automatic correction method for a binocular vision system provided by Embodiment 1 of the present invention.

[0037] The method includes:

[0038] Step 1: Acquire an RGBD image from a front-facing stereo camera mounted on a mobile device. The mobile device is, for example, a car, and the forward-facing stereo device is a multi-eye camera, which includes at least two cameras, namely a left camera and a right camera. During the driving of the car, the left camera and the right camera continuously capture images, according to the left camera and the right camera. Two corresponding images captured by the right camera at the same time can be used for depth recognition to form an RGBD image. The specific processing method is the prior art in the field, and this patent will not describe it in detail for the purpose of simplification.

[0039] Step 2: Estimate the height of the road surface using the RGBD image.

[0040] Step 3: For obstac...

Embodiment 2

[0052] Figure 4 It is a block diagram of an automatic correction system of a binocular vision system provided by Embodiment 2 of the present invention.

[0053] The method includes:

[0054] Apparatus 1 is used to acquire an RGBD image from a front-facing stereo camera installed on a mobile device. The mobile device is, for example, a car, and the forward-facing stereo device is a multi-eye camera, which includes at least two cameras, namely a left camera and a right camera. During the driving of the car, the left camera and the right camera continuously capture images, according to the left camera and the right camera. Two corresponding images captured by the right camera at the same time can be used for depth recognition to form an RGBD image. The specific processing method is the prior art in the field, and this patent will not describe it in detail for the purpose of simplification.

[0055] The device 2 is used for estimating the road surface height by using the RGBD ...

Embodiment 3

[0068] A computer program for performing the following methods, comprising:

[0069] Step 1: Acquire an RGBD image from a front-facing stereo camera mounted on a mobile device. The mobile device is, for example, a car, and the forward-facing stereo device is a multi-eye camera, which includes at least two cameras, namely a left camera and a right camera. During the driving of the car, the left camera and the right camera continuously capture images, according to the left camera and the right camera. Two corresponding images captured by the right camera at the same time can be used for depth recognition to form an RGBD image. The specific processing method is the prior art in the field, and this patent will not describe it in detail for the purpose of simplification.

[0070] Step 2: Estimate the height of the road surface using the RGBD image.

[0071] Step 3: For obstacles higher than the road surface, calculate the slope of the obstacle.

[0072] Step 4: For the area whos...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses an exercisable region detection method based on RGBD (Red, Green and Blue Depth). The method comprises the following steps that: S1: obtaining an RGBD image from forward direction stereo camera equipment installed on movable equipment; S2:utilizing the RGBD image to estimate pavement height; S3: for an obstacle which is higher a pavement, calculating the gradient of the obstacle; and S4: for an area of which the height is greater than a preset value and the gradient is greater than a preset deserved area as a non-driving area.

Description

Background technique [0001] In the field of automatic driving or assisted driving, computer vision is used to analyze images captured by multi-eye cameras, and parallax analysis and image recognition are used to determine which areas are road surfaces and which areas are obstacles, so as to determine the driving area. [0002] In the prior art, it mainly uses computer learning technology to learn the features of road surface images in advance and save these features in the feature library. Come on the road. [0003] There are many problems in this method, such as the ever-changing road surface images, it is difficult to cover all through prior learning. And, to identify the graphical elements in the logical sense of the image, the amount of calculation is very large, and it is difficult to achieve the timeliness of identification. [0004] There is an urgent need in the field for a technology that can quickly and efficiently identify drivable areas. Contents of the inventi...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/00G01C9/00G01C11/04
CPCG01C9/00G01C11/04G06V20/58
Inventor 郑继川
Owner 元橡科技(北京)有限公司