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

Digital liquid crystal-oriented segmentation and identification method

A recognition method and digital technology, applied in character and pattern recognition, instruments, biological neural network models, etc., can solve problems such as poor detection results, and achieve the effect of reducing background noise and accurate digital segmentation

Inactive Publication Date: 2019-08-30
NANJING UNIV +1
View PDF8 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Image processing technology can perform image enhancement and feature point matching, and SIFT feature point matching can detect and locate objects, but the detection effect is not good for pictures with complex backgrounds or objects without obvious features, so it is necessary to preprocess the pictures to remove interference

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
  • Digital liquid crystal-oriented segmentation and identification method
  • Digital liquid crystal-oriented segmentation and identification method
  • Digital liquid crystal-oriented segmentation and identification method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0027] figure 1 The overall process of the present invention is given. The present invention mainly includes dial positioning, number segmentation and number recognition.

[0028] (1) Dial positioning is to find the position of the LCD dial on the input image, and cut out the picture of the dial, and specify the specific implementation steps of dial positioning:

[0029] (1.1) Prestore different types of digital LCD meter template pictures, and intercept the area containing the entire dial. In order to obtain multiple stable feature points, keep the corner area as much as possible when intercepting; mark the coordinates of the display area on the dial template picture , Save as json file format;

[0030] (1.2) Use the Sift algorithm to obtain the feature matrix of the input image and the feature matrix of the corresponding template picture, calculate the feature distance to obtain the position of the dial, and cut out the dial from the input image;

[0031] (2) Digital segmentation in...

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 novel digital liquid crystal-oriented segmentation and identification method. The invention defines a new calibration method which can be applied to the situation that the display screen has stains and scratches, and the digital table is accurately segmented through calibration. According to the method, the problem of no liquid crystal digital set is solved, namely the existing digital set is preprocessed to be consistent with the to-be-processed digital set in distribution. The invention provides a convolutional neural network structure which is suitable for identification of small-size and small-order-of-magnitude digital pictures.

Description

Technical field [0001] The invention relates to a new segmentation and recognition method for digital liquid crystals. Background technique [0002] Computer vision has many applications in target detection and recognition. It can be divided into two categories: digital image processing technology and convolutional neural network. Digital image processing technology can detect geometric figures or simple objects without samples or a small number of samples. With the popularity of deep learning methods, especially convolutional neural networks, they can extract higher semantic features of images, but require a large number of samples. The neural network model has better generalization performance, and the recognition effect is better when the training set is sufficient. . [0003] The automatic identification of LCD digital meters has high application value in the industrial field, but accurate segmentation and identification of numbers are difficult. Different LCD digital meters...

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): G06K9/32G06K9/46G06K9/62G06N3/04
CPCG06V10/25G06V10/462G06V20/625G06N3/045G06F18/214
Inventor 徐园园黄继鹏高阳
Owner NANJING UNIV
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