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

Road lane line detection method and system based on deep neural network

A deep neural network and lane line detection technology, applied in the field of lane line detection, can solve the problem of reducing the network model and achieve the effect of improving the accuracy of prediction, saving computing resources, and improving stability

Active Publication Date: 2022-06-07
HUNAN UNIV
View PDF10 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0007] The present invention provides a road lane line detection method and system based on a deep neural network, which is used to solve the problem that the lane line detection method needs to reduce the parameters and computational complexity of the network model, and consider that the temporally continuous image frames of the driving road scene have Technical issues associated with a priori properties

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 lane line detection method and system based on deep neural network
  • Road lane line detection method and system based on deep neural network
  • Road lane line detection method and system based on deep neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0036] Embodiment 1, a road lane line detection method based on a deep neural network.

[0037] This embodiment provides a method for detecting road lane lines based on a deep neural network, including the following steps:

[0038] Acquire the image to be recognized and the N continuous images before and after it, and form a set of continuous image sequences according to the time sequence;

[0039] The continuous image sequence is input into the pre-trained deep neural network model; the deep neural network model includes an encoding network module, a recurrent convolutional network module that explores and learns temporal prior information, and a decoding network module;

[0040] Through the coding network module, feature extraction is performed on each image in the continuous image sequence in turn, and the feature map sequence containing the semantic features of the lane line is obtained;

[0041] The cyclic convolutional network module receives the cyclic input of the fea...

Embodiment 2

[0051] Embodiment 2, a computer system.

[0052] This embodiment provides a computer system, including a memory, a processor, and a computer program stored in the memory and running on the processor. When the processor executes the computer program, the deep neural network-based road lane described in Embodiment 1 is implemented. Steps of the line detection method.

Embodiment 3

[0053] Embodiment 3, a computer-readable storage medium.

[0054] This embodiment provides a computer-readable storage medium, where the computer-readable storage medium stores computer program instructions, wherein, when the computer program instructions are executed by the processor, the processor causes the processor to execute the deep neural network-based road described in Embodiment 1 The steps of the lane line detection method.

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 present invention provides a lane line detection method and system based on a deep neural network. The method performs feature extraction on continuous images through an encoding network module to obtain a sequence of feature maps containing semantic features of lane lines, and then inputs the sequence of feature maps to the corresponding In the circular convolutional network module of the network module, multi-layer circular convolution and temporal feature fusion are performed on the feature map sequence through the circular convolutional network module, and the semantic feature map after the feature fusion is output, and finally the semantic feature is processed by the decoding network module. The map is decoded, and the predicted map of the position of the lane line is output. The present invention fully considers the continuity of the driving scene, that is, the association on the time series, and can learn the time prior association information, thereby improving the stability of the network and the accuracy of prediction, and can also combine the previous images in the continuous images The feature information of the image is used in the post-stage image, which can reduce the parameters and computational complexity of the network model, and save computational resources.

Description

technical field [0001] The invention relates to the technical field of lane line detection in automatic driving assistance, in particular to a road lane line detection method and system based on a deep neural network. Background technique [0002] With the rapid development of technology and economic level of the automobile industry in today's society, the number of motor vehicles in the world has risen rapidly. The widespread popularity of automobiles brings convenience to people's lives, but also causes many problems to people's life safety, such as traffic congestion, traffic accidents, and environmental pollution. With the increase of vehicle traffic accidents, road safety has become more and more important, and the automatic driving and advanced assisted driving functions of automobiles have received more and more attention and research. In autonomous driving and advanced driving assistance systems, vision-based lane line detection is the core of lane departure warning...

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): G06V20/56G06V10/80G06V10/82G06N3/04G06N3/08
CPCG06N3/08G06N3/045G06F18/253
Inventor 肖德贵卓林
Owner HUNAN UNIV