Method and device for realizing handwritten numeral recognition based on graphic edge detection

An edge detection and digital recognition technology, applied in the field of intelligent image recognition, can solve problems such as multiple digital samples, complex accuracy of deep learning models, and difficulty in having high guarantees, so as to improve authenticity, shorten project development cycles, and enhance product competition. force effect

Inactive Publication Date: 2020-11-27
CHENGDU SEFON SOFTWARE CO LTD
View PDF1 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to provide a method and device for realizing handwritten digit recognition based on graphic edge detection, so as to solve the problem that the existi

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
  • Method and device for realizing handwritten numeral recognition based on graphic edge detection
  • Method and device for realizing handwritten numeral recognition based on graphic edge detection
  • Method and device for realizing handwritten numeral recognition based on graphic edge detection

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0053] A method for realizing handwritten digit recognition based on graphic edge detection includes the following steps: first, preprocessing the hand drawing to obtain a binary image; then extracting the outline of a single digit in the binary image; and then calculating the outer frame of a single digit Arrange the outer frame in ascending order; finally identify the numbers in the outer frame, and then connect the individual numbers to form the overall number;

[0054] In the present invention, after binarizing the hand-drawn drawing, the contour line of a single number is extracted, and then the frame of the single number is calculated and sorted, and finally the single numbers are connected to form an overall number; Then compare with the data model that contains a large number of abundant training samples to make the final recognition judgment. The present invention is aimed at a single number, and all integral numbers are composed of 10 numbers from 0 to 9, so the sample ...

Embodiment 2

[0057] In this embodiment, on the basis of Embodiment 1, the OpenCV library is used to grayscale and binarize the hand-drawn image to obtain a binarized image with only two values ​​of 0 and 255 for the entire image.

Embodiment 3

[0059] In this embodiment, on the basis of Embodiment 1, the specific process of extracting the contour line of a single number in the binarized image is as follows: first use the edge extraction algorithm to extract the contour line of a single number in the binarized image; then delete step S201 Boundary contour lines in contour lines; then delete contour lines in contour lines whose contour area is smaller than the set threshold; finally get the outer contour line or outer contour line and inner contour line of a single number; if the number "1" only has the outer contour, the number "0" has both outer and inner contours.

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 method and a device for realizing handwritten numeral recognition based on graph edge detection, and mainly solves the problems that more numeral samples are required, a deeplearning model is complex and the accuracy is difficult to guarantee in the existing deep learning OCR handwritten numeral recognition technology in the prior art. The method comprises the followingsteps of: preprocessing a hand drawing to obtain a binary image; then extracting a contour line of a single number in the binary image; calculating an outer frame of a single number and then carryingout ascending sort on the outer frame; and finally, identifying the numbers in the outer frame, and connecting the single numbers to form an integral number. Through the scheme, the purposes of simplemodel and high recognition accuracy are achieved, and the method has very high practical value and promotional value.

Description

technical field [0001] The invention relates to the field of intelligent image recognition, in particular to a method and device for recognizing handwritten digits based on image edge detection. Background technique [0002] Handwritten digit recognition has always been a research hotspot in the field of pattern recognition and machine learning. This technology has been widely used in tax form processing, mail sorting and bank check recognition. In these applications, handwritten digit recognition algorithms are usually required to have high recognition speed and recognition accuracy as well as high reliability and stability. [0003] Deep learning OCR handwritten digit recognition technology uses deep learning technology to construct a digital recognition model through a large number of training samples composed of handwritten digits to recognize digits at one time, but deep learning OCR handwritten digit recognition technology is to recognize the entire digit at one time, ...

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
IPC IPC(8): G06K9/00G06K9/20G06K9/34G06K9/38G06T7/11G06T7/13G06T7/136G06K9/46G06K9/62G06N3/04G06N3/08
CPCG06T7/11G06T7/13G06T7/136G06N3/08G06V40/33G06V10/22G06V10/267G06V10/28G06V10/44G06N3/045G06F18/24
Inventor 徐泽鸿韩威宏张艳王国玉
Owner CHENGDU SEFON SOFTWARE CO LTD
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