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

A visual image detection algorithm based on grille background

A visual image detection and plate technology, applied in image enhancement, image analysis, image data processing and other directions, can solve the problems of variable color, low grid background recognition accuracy, and inability to adapt to rich textures, etc., to achieve easy implementation and simple principle. Effect

Inactive Publication Date: 2018-12-25
浙江优迈德智能装备有限公司
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to solve the shortcomings of existing recognition methods that cannot adapt to the grid background with rich texture and variable color or low recognition accuracy, and propose a visual detection method for plates under the grid background

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
  • A visual image detection algorithm based on grille background
  • A visual image detection algorithm based on grille background
  • A visual image detection algorithm based on grille background

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0031] First read the plate to get the image I; then extract the texture in the image I to get the texture image I t ; Then to the texture image I t Perform image morphology operations to obtain image I 0 ; then in I 0 Contour detection is performed in the image, and all the contours in the image are read; then the contours are screened according to the area, shape and other information; then if the plate is a rectangle, the smallest enclosing rectangle of its contour is taken as its final contour, otherwise the filtered contour is used as the final outline.

Embodiment 2

[0033] In embodiment one, add following operation:

[0034] The texture extraction in step (2) uses an edge detection operator for extraction, and the edge detection operator uses one of the Sobel operator and the Canny operator, or uses the convolution kernel with the same local structure as the grid image to repeat the image. Convolution operation, and then binarize the image to extract.

[0035] First read the plate to get the image I; then extract the texture in the image I to get the texture image I t , where the texture extraction uses the edge detection operator for extraction, the edge measurement operator uses one of the Sobel operator and the Canny operator, or uses the same convolution kernel as the repeated local structure of the grid image to perform convolution operations with the image, Then the image is binarized to extract; then the texture image I t Perform image morphology operations to obtain image I 0 ; then in I 0 Contour detection is performed in the...

Embodiment 3

[0037] In embodiment two, add following operation:

[0038] The texture image in step (2) is a binary image, where the texture is white and the rest is black, or the texture is black and the rest is white.

[0039] First read the plate to get the image I; then extract the texture in the image I to get the texture image I t , where the texture extraction uses the edge detection operator for extraction, the edge measurement operator uses one of the Sobel operator and the Canny operator, or uses the same convolution kernel as the repeated local structure of the grid image to perform convolution operations with the image, Then the image binarization is extracted, the texture image is a binary image, the texture is white, and the rest of the positions are black, or the texture is black, and the rest of the positions are white; then the texture image I t Perform image morphology operations to obtain image I 0 ; then in I 0 Contour detection is performed in the image, and all the ...

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 plate visual image detection algorithm under the grille background, comprising the following steps: (1) reading the plate to obtain an image I; (2) extracting texture from image I to obtain texture image It; 3) carrying out image morphological operation on the texture image It to obtain an image I0; (4) contour detection being carried out in I0, and all contours in the image being read out; (5) screening the contour according to the information of area and shape; (6) if the plate is a rectangle, taking the smallest surrounding rectangle of its contour as its final contour, otherwise taking the contour in step (5) as the final contour. The method detects the plate area in the image by utilizing the features of less plate texture and more grille background texture,the principle is simple and easy to realize. For the plate with less texture, the method has better effect, can quickly find the location of the plate, and solves the problem of low detection efficiency by using various image filtering, morphological operation and other methods.

Description

technical field [0001] The invention relates to the technical field of machine vision, in particular to a plate visual image detection algorithm under a grid background. Background technique [0002] In automatic processing such as automatic painting of plates, it is often necessary to detect the area of ​​the plate from the grid on which the plate is placed by machine vision. The existing recognition methods are difficult to adapt to this situation. Specifically, the method of contour extraction on the texture image requires a simple background with few textures, and the grid background is composed of a large number of textures, so it is not suitable; it is recognized by color The object method requires that the object to be recognized has an obvious color difference from the background, and in a painting environment, the color of the object and the background will change as the painting progresses. For objects that need to be painted multiple times, the color of the object...

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): G06T7/00G06T7/13G06T7/49
CPCG06T7/0004G06T7/13G06T7/49G06T2207/20036
Inventor 于兴虎刘伟良卫作龙李湛佟明斯林伟阳
Owner 浙江优迈德智能装备有限公司
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