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

Monocular vision road recognition algorithm based on appearance and geometric information fusion

A technology of geometric information and road recognition, applied in the field of computer vision, can solve problems that affect the stereo matching effect of binocular vision, affect the recognition accuracy, and have repeated textures, etc.

Active Publication Date: 2018-04-20
ZHEJIANG UNIV
View PDF5 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the road area in the actual situation is often weak in texture or has repeated textures, which greatly affects the effect of binocular vision stereo matching, which in turn affects the recognition accuracy

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
  • Monocular vision road recognition algorithm based on appearance and geometric information fusion
  • Monocular vision road recognition algorithm based on appearance and geometric information fusion
  • Monocular vision road recognition algorithm based on appearance and geometric information fusion

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0080] The present invention will be further described below in conjunction with drawings and embodiments.

[0081] Embodiments of the present invention are as follows:

[0082] 1) Initialize the Gaussian mixture model and model the appearance information of the road area using the original image of k frames and the illumination-invariant image, and then use the Gaussian mixture model in k+1 frames to estimate the road area of ​​the current frame to obtain the current frame The probability value of each pixel in the original image and the illumination-invariant image belongs to the road area. Process such as image 3 as shown, image 3 The left column uses the original image of k frames and the illumination invariant image to initialize the Gaussian mixture model, the right column uses the Gaussian mixture model to estimate the road area in k+1 frame, and the last line is the road area of ​​k frame and k+1 frame Estimated results. It can be seen from the comparison that th...

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 a monocular vision road recognition algorithm based on appearance and geometric information fusion. According to the algorithm, images of road scenes in front of an automobileare shot and collected in real time through a camera installed on the automobile, and the images collected in real time are sequenced; appearance information of the images is acquired; geometric information of the images is acquired; the appearance information and the geometric information are utilized to perform road pre-partitioning processing; and finally road edge optimization is performed. Through the method, the image information and the geometric information are fully utilized to recognize a road region, the precision and reliability of road recognition are improved, and the precision and geometric smoothness of road recognition in a three-dimensional space are guaranteed.

Description

technical field [0001] The invention belongs to the field of computer vision and relates to a monocular vision road recognition algorithm based on fusion of appearance and geometric information. Background technique [0002] During the operation of robots and intelligent vehicles, it is necessary to identify drivable road areas to improve their driving safety. However, in the outdoor environment, the effect of the visual algorithm is affected by complex environmental factors, which increases the difficulty of the recognition task and the complexity of the visual recognition algorithm. [0003] In previous studies, the appearance information of the image or the geometric information of the scene is often used to identify the road area. Methods based on image appearance information generally use the color and texture features of image pixels to classify pixels (J. Alvarez, T. Gevers, Y. LeCun, A. Lopez. Road scenes segmentation from a single image. European Conference on Comp...

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): G06K9/00G06K9/62
CPCG06V20/588G06F18/23G06F18/22
Inventor 陈剑贾丙西王麒张凯祥
Owner ZHEJIANG UNIV