Road Edge Detection Method Based on Heuristic Probabilistic Hough Transform

A heuristic, probabilistic technology, applied in character and pattern recognition, image data processing, instrumentation, etc., to solve problems such as wrong road detection results

Active Publication Date: 2017-06-20
NANTONG QIWU AGRI PROD CO LTD
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] In the mapping stage from the image space to the parameter space, the traditional PHT method randomly selects candidate edge points from the image space without using the previous heuristic information (detection results such as the angle and position of the road boundary); the RHT method uses the previous road detection The information divides the target search area OSA, but makes the edge points outside the OSA have no chance to participate in the mapping. When the road boundary is outside the OSA due to sudden changes, it may lead to wrong road detection results.

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 Edge Detection Method Based on Heuristic Probabilistic Hough Transform
  • Road Edge Detection Method Based on Heuristic Probabilistic Hough Transform
  • Road Edge Detection Method Based on Heuristic Probabilistic Hough Transform

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0055] A road-weighted Canny edge detection method. For each edge point subset, the least square method is used to fit the straight line first, and the axis of the point set is obtained. Suppose the equation of the straight line is L: y=kx+b, then the distance from any point (x, y) in the subset to the straight line is Foot coordinates ((ky+x-kb) / (1+k 2 ),(k 2 y+kx+b) / (1+k 2 )). Suppose two points P in the subset 1 ,P 2 Perpendicular point P on line L v1 ,P v2 are the two endpoints of all the vertical points on the axis, it can be easily proved that |P 1 P 2 |≥|P v1 P v2 |Established. So, P 1 ,P 2 to P v1 ,P v2 The mapping of is a compressed mapping, which means that the line segment P v1 P v2 For any foothold point on , at least one point corresponding to it can be found in the edge point subset. Select the point with the smallest distance in the subset of edge points as its corresponding point, and other points with the same vertical feet as non-correspond...

Embodiment 2

[0060] A heuristic probabilistic Hough transform method for weighted Canny edge images. According to the historical recognition results of the road boundary line L, the entire image space is divided into regions of interest and regions of non-interest. Edge points distributed in different regions have different probabilities value. as attached figure 1 , 2 As shown, the shaded area is the ROI area, and the edge points in this area have a higher probability value P2, so the probability of being randomly selected is higher, while the probability value of edge points in other areas is P1, and the probability of being selected lower. This means that the longer edge line segment in the ROI area will have a high chance to be selected first, rather than the longest line in the image space. Although the candidate edge points in the non-ROI area have a low probability of being selected, they do not deprive them of the opportunity of random mapping. When the road boundary suddenly oc...

Embodiment 3

[0069] A road weighted Canny edge bi-polyline model extraction method. The bi-polyline model divides the road boundary into two sections: the near section and the far section, which represent the direction and trend of the near road and the far road respectively. The extraction method of the model is as follows:

[0070] 1. Extract the weighted Canny edge from the original image, and filter according to the weight of the image edge to generate an edge family of candidate road boundaries.

[0071] 2. Use the least square method to estimate the main direction of the candidate edge family, and project all the edge points on the main axis generated by the least square method, and select the center position O of the edge family on the main axis according to the projection range.

[0072] 3. A straight line passing through point O and perpendicular to the main axis divides the edge family into two parts. First, Hough transform is performed on the nearby edge family, and then the ent...

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 relates to a road edge detection method based on heuristic probability Hough transform. As an adaptive threshold search method, the Otsu method is robust to unstructured road region segmentation in complex environments. However, the Otsu method itself can only obtain rough segmentation results of road areas and non-road areas, but cannot obtain precise road boundaries. The method of the present invention comprises the following steps: in the double-threshold detection stage, the edge pixels connected to each other are uniformly numbered, thereby forming several subsets of the Canny edge; first, the least squares method is used for each edge point subset to carry out straight line fitting to obtain The axis of this point set. The invention is used for road edge detection.

Description

[0001] Technical field: [0002] The invention relates to a road edge detection method based on heuristic probability Hough transform. [0003] Background technique: [0004] Otsu, as an adaptive threshold search method, is robust to unstructured road region segmentation in complex environments. However, the Otsu method itself can only obtain rough segmentation results of road areas and non-road areas, but cannot obtain precise road boundaries. How to make the intelligent vehicle know whether there is a passable area ahead, how to control the intelligent vehicle to drive along the direction of the road, etc., all of which depend on the detection of the roadside. While driving outdoors, road conditions and ambient light have a great influence on the results of roadside detection. Whether the edge detection algorithm can be disturbed by these factors as little as possible. On the other hand, in order to control the driving of intelligent mobile vehicles in real time, there are ...

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): G06T7/13G06K9/46
Inventor 王燕清石朝侠陈德运孙晓君孙广路李扬李松唐远新
Owner NANTONG QIWU AGRI PROD CO LTD
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

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

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap