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

Screen image recognition method and equipment, system and readable storage medium

A screen image and equipment technology, applied in the field of computer vision, can solve the problems of insufficient stability, low response speed, and poor applicability of digital image processing technology, and achieve the goals of improving adaptability and intelligence, fast response ability, and simplifying complexity Effect

Active Publication Date: 2020-12-18
敬科(深圳)机器人科技有限公司
View PDF17 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] However, for some complex scenes or large external interference, the stability of the above-mentioned digital image processing technology is insufficient, and the response speed is low, resulting in poor applicability

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
  • Screen image recognition method and equipment, system and readable storage medium
  • Screen image recognition method and equipment, system and readable storage medium
  • Screen image recognition method and equipment, system and readable storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0055] The embodiment of the present application discloses a screen image recognition method, referring to figure 1 , the method consists of,

[0056] 101. Acquire multiple reference images containing a target scene.

[0057] 102. Extract target regions in multiple reference images containing target scenes to obtain multiple reference templates.

[0058] 103. Extract multiple feature points of multiple reference templates.

[0059] 104. Train the neural network according to multiple feature points of multiple reference templates.

[0060] 105. Extract multiple feature points of the target image.

[0061] 106. Perform geometric registration calculation on multiple feature points of the target image and multiple feature points of multiple reference templates to obtain a similarity.

[0062] Specifically, a feature map of the target image is acquired according to multiple feature points of the target image.

[0063]According to the multiple feature points of the multiple ref...

Embodiment 2

[0068] The embodiment of the present application discloses a screen image recognition method, referring to figure 2 , the method includes:

[0069] 201. Acquire multiple reference images containing a target scene.

[0070] Specifically, the initial sample collection is performed on the liquid crystal display screen interface including the target scene, and stored in a temporary sample library as a reference image.

[0071] For different types of screens, the target scene may be a scene of a certain type of screen;

[0072] In practical applications, multiple different types of MLDs can be

[0073] Execute the operation described in step 201 for multiple target scenes respectively corresponding to the type screens.

[0074] Optionally, in practical applications, the process described in step 201 may be realized by shooting the target scene with a camera, and the process may specifically be:

[0075] Setting the exposure time, wherein the exposure time is less than the firs...

Embodiment 3

[0140] The embodiment of the present application discloses a screen image recognition device, refer to image 3 , the device consists of:

[0141] Reference extraction module 31, for extracting a plurality of feature points to a plurality of reference templates;

[0142] Target extraction module 32, for extracting a plurality of feature points to target image;

[0143] Registration calculation module 33: for performing geometric registration calculation on the plurality of feature points of the target image and the plurality of feature points of the plurality of reference templates to obtain similarity;

[0144] The first identification module 34: if the similarity is greater than or equal to the similarity threshold, output a first identification result according to the reference template;

[0145] The second identification module 35: if the similarity is smaller than the similarity threshold, identify multiple feature points of the target image through a neural network, an...

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 screen image recognition method and equipment, a system and a readable storage medium. The method comprises the steps of extracting multiple feature points of a target image; performing geometric registration calculation on the plurality of feature points of the target image and the plurality of feature points of the reference template to obtain similarity; if the similarity is greater than or equal to a similarity threshold, outputting a first identification result according to the reference template; and if the similarity is smaller than the similarity threshold, identifying the feature points of the target image through a neural network, and outputting a second identification result. According to the method, the stability and the response speed of screen imagerecognition are improved, and the applicability of screen image recognition is further improved.

Description

technical field [0001] The present application relates to the technical field of computer vision, in particular to a screen image recognition method, device, system and readable storage medium. Background technique [0002] In the era of rapid development of computer technology, labor costs are also rising with various factors, and computer vision has gradually penetrated into various fields of industrial production. In the automation industry, computer vision is a promising development direction. It can make machines have a visual system similar to eyes and brains like humans, and can complete various complex tasks. [0003] For example, in some low-level automation scenarios, the machine is given a fixed program, repeats the same action or function, does not have the ability to self-perceive the target, and in some cases cannot completely replace human labor, such as cars. Spraying, or in the field of industrial inspection, for some parts with shape defects, it often requ...

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): G06K9/62G06T7/33G06N3/08
CPCG06T7/33G06N3/08G06F18/22
Inventor 陈辉孙敬颋高会军林伟阳崔鹏
Owner 敬科(深圳)机器人科技有限公司
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