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

Ancient character and font recognition method based on improved YOLO v3

A font recognition and text technology, applied in the field of ancient text image recognition based on improved YOLOv3, can solve problems such as unrecognizable fonts, low recognition accuracy, and difficult ancient text recognition

Pending Publication Date: 2020-05-08
HANGZHOU DIANZI UNIV
View PDF3 Cites 12 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

At present, there are mostly modern Chinese characters or handwritten Chinese character recognition, and optical character recognition. The former has a low recognition accuracy and is difficult to apply to ancient character recognition. The latter can be used for ancient Chinese character recognition but cannot recognize its fonts.

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
  • Ancient character and font recognition method based on improved YOLO v3
  • Ancient character and font recognition method based on improved YOLO v3
  • Ancient character and font recognition method based on improved YOLO v3

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0045] Attached below Figure 5 The present invention will be described in further detail with reference to specific embodiments.

[0046] 1) According to whether the amount of original data of each character can meet the requirements of neural network training and whether it belongs to commonly used ancient Chinese characters, ancient characters with more than 30 original pictures are selected as the data set. The entire data set contains 100 characters in three fonts: oracle bone inscriptions, bronze inscriptions and bamboo slips and silk from Chu, and a total of 4,000 original pictures, such as figure 2 shown;

[0047] 2) Expanding the collected ancient text images to obtain ancient text sample images;

[0048] In the step 2), data enhancement is used. Generally speaking, the parameters of the neural network are in the millions, and a large amount of data is required for training to obtain parameters that can work correctly. In order to increase the amount of training d...

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 an ancient character and font recognition method based on improved YOLO v3. The method is a target detection method based on deep learning. A deep neural network structure is utilized to learn an overall-partial decomposition relationship in an ancient character image, useful feature information is obtained through a feature extraction network, detection and positioning areperformed, then the feature information of the image is sent to a classifier for classification and recognition, and the position of the ancient character is framed in the image by using a bounding box. According to the method, the problems that ancient characters have complex internal structures and the precision is low when the characteristics are used for recognition are solved. According to the improved YOLO v3 provided by the invention, the ShuffleNet v2 is used as a main structure of the model, so that the improved YOLO v3 is more efficient. The accuracy of ancient character and font recognition reaches 98.81%, and the method has good stability and good robustness and can be applied to ancient character recognition scenes such as ancient character texts and inscription stickers.

Description

technical field [0001] The present invention relates to an image recognition method based on a deep learning target detection algorithm, in particular to an image recognition method for ancient characters based on improved YOLO v3. Background technique [0002] Ancient characters record the social life of ancient people, and the study of ancient characters serves as a key to open the convenient door to study the lives of ancient people. Paleography plays an important role in the study of ancient Chinese history and culture. The written content on some physical materials such as oracle bones, bronze wares, stone tablets, and ancient books contains a lot of important historical information. Interpretation of these ancient texts is helpful for understanding the social conditions at that time. However, ancient text images are very complex, with rich and logical structural information. Ancient characters have a complex internal structure, many strokes, complex strokes, high sim...

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): G06K9/34G06K9/62G06T3/40G06T5/00G06N3/04G06N3/08
CPCG06T3/4007G06N3/08G06T2207/10004G06T2207/20081G06T2207/20084G06V30/153G06V30/10G06N3/045G06F18/2431G06F18/253G06T5/70Y02D10/00
Inventor 董哲康石杰高明煜齐冬莲林辉品吴俊洁
Owner HANGZHOU DIANZI 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

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