Image distortion correction method based on bp neural network for fisheye lens

A BP neural network and fisheye lens technology, applied in the field of fisheye lens image distortion correction, can solve the problems of time-consuming and labor costs, and achieve the effect of improved efficiency and strong self-learning

Active Publication Date: 2020-04-24
SYSU CMU SHUNDE INT JOINT RES INST +1
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] In order to solve the defects that the above image distortion correction method needs to consume a lot of time and labor costs, the present invention provides a method for correcting distortion of images captured by a fisheye lens based on BP neural network, and the efficiency of this method for correcting distortion is compared with the existing got improved

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
  • Image distortion correction method based on bp neural network for fisheye lens
  • Image distortion correction method based on bp neural network for fisheye lens
  • Image distortion correction method based on bp neural network for fisheye lens

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0051] Such as Figure 18 As shown, the fisheye lens image distortion correction method provided by the present invention includes the following steps;

[0052] S1. If figure 1 As shown, assume that there are m rows and n columns of feature points evenly distributed on the paper A, use the fisheye lens to shoot the paper A, and obtain the captured image, as shown in figure 2 shown;

[0053] S2. Preprocessing the image, the obtained image is as follows image 3 , Figure 13 as shown in (a);

[0054] S3. Extract feature points from the preprocessed image, specifically as Figure 4 , Figure 13 as shown in (b);

[0055] S4. Among the extracted feature points, the horizontal distance x between the two adjacent feature points with the largest horizontal distance is the distance between the feature points in each column, and the vertical distance y between the two adjacent feature points with the largest vertical distance is The distance between feature points in each row; ...

Embodiment 2

[0071] This embodiment has carried out concrete simulation experiment to the method provided by the present invention, and its experiment process and result are as follows Figure 14~17 As shown, it can be seen that some of the input feature points cannot be well recognized due to the edge relationship, and some feature points are lost in the isolated point extraction. After the correction of the neural network, a better correction effect can still be obtained.

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 provides a BP neural network-based image distortion correction method for an image shot by a fish-eye lens. Thus, compared with the prior art, the efficiency of distortion correction is improved. Because the neural network is characterized by strong self-learning ability, self-organization, high nonlinearity and high robustness, the method has specific advantages in solving problems like nonlinearity fitting, and multi-layered complex problems can be solved. According to the invention, by use of the neural network to solve image distortion correction problem, constrains of the traditional image distortion correction technology are broken through and the method is advantaged by irreplaceability in the aspect of nonlinearity distortion correction.

Description

technical field [0001] The invention relates to the field of neural networks, and more specifically, to a method for correcting distortion of images captured by a fisheye lens based on a BP neural network. Background technique [0002] With the development of intelligent science, computer vision technology, and people's growing demand for video surveillance and acquisition, fisheye lenses with a wide field of view are more and more widely used in daily life. However, the photos and videos obtained by using the fisheye lens will have serious distortion. In order to obtain images that people are accustomed to and images that can be recognized by computers, it is necessary to perform distortion correction processing on images obtained through fisheye lenses. The traditional distortion correction method needs to build a model of the image before and after correction according to the parameters of the lens. Since the parameters of different lenses are different, it takes a lot o...

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): G06T5/00G06N3/04
CPCG06N3/04G06T5/006
Inventor 王军谢启超陈谋奇
Owner SYSU CMU SHUNDE INT JOINT RES INST
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