Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Road detection method based on monocular vision

A road detection and monocular vision technology, applied in the fields of computer vision and machine learning, can solve the problem of low detection accuracy, and achieve the effect of enhancing classification accuracy and road detection accuracy.

Active Publication Date: 2020-11-17
XI'AN INST OF OPTICS & FINE MECHANICS - CHINESE ACAD OF SCI
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The disadvantage of this method is that the default assumption is that the road boundary can be fitted by two rays at the same vertex, and the detection accuracy is low when the road is obviously curved.

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
  • Road detection method based on monocular vision
  • Road detection method based on monocular vision
  • Road detection method based on monocular vision

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0028] Combine below figure 1 The present invention is further described.

[0029] The invention proposes a road detection method based on monocular vision online learning. The method obtains the structure distribution of the input data by using an online structured support vector machine (SSVM), and at the same time updates the classifier online so that it can adapt to environmental changes. to combine figure 1 , input video, I={I t |t=1,2...,n}, n is the number of video frames. The realization steps of the present invention are as follows:

[0030] Step 1, for the input monocular image, judge whether it is the first frame, if it is the first frame, go to step 2, otherwise go to step 3.

[0031] Step 2, initial input data I 1 deal with.

[0032] (2a) For the first input frame data I 1 , normalize the image to 300×500 size. The positive samples use the manual labeling method to determine the location of the road edge, and use the binary image R 1 Describe the locatio...

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-based road detection method. The method mainly comprises steps: (1) road edge area positive and negative sample sampling is carried out on a first frame of image, local feature descriptors are extracted in the selected positive and negative samples, and a structural support vector machine is used as a classifier for training to obtain a classification plane for an initial frame of sample; and (2) as for a subsequent test frame of image, road area position in the former frame is used for determining a candidate sample sampling area for the test frame, and the structural support vector machine obtained through training of the former frame is used for determining the road attribution of the sample in the test frame. Meanwhile, a random sample consensus method is adopted to use two intersected lines to fit a road edge position, and the road area is determined finally. Thus, the accurate position of the road can be obtained, and effective prior information is provided for automotive assistant driving, pedestrian vehicle detection and the like.

Description

technical field [0001] The invention belongs to the technical field of computer vision and machine learning, in particular to an online road detection method based on monocular vision. Background technique [0002] In recent years, with the continuous increase of the road mileage and the number of cars, the casualties and property losses caused by traffic accidents are also increasing rapidly. In 2012, the number of deaths due to traffic accidents in my country reached 59,000, and the direct property loss was 1.17 billion yuan. In this context, automotive assisted driving systems, including lane departure warning, pedestrian warning, anomaly detection, etc., can effectively improve car driving safety and reduce casualties and property losses. A key technology of the system is road detection, because road detection can provide key constraint information for the car's assisted driving system. [0003] According to the types of sensors used, road detection technology can be d...

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 Patents(China)
IPC IPC(8): G06K9/00G06K9/66
Inventor 袁媛王琦姜志宇
Owner XI'AN INST OF OPTICS & FINE MECHANICS - CHINESE ACAD OF SCI
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com