Irregular text correction and recognition method and system based on saliency map
A recognition method and text recognition technology, applied in the field of text recognition in natural scenes, can solve problems such as low recognition accuracy and inability to recognize irregular text well, and achieve the effect of improving accuracy and robustness
Pending Publication Date: 2020-05-12
NANJING UNIV +1
View PDF2 Cites 5 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] The technical problem to be solved by the present invention is to overcome the defects that the existing irregular text recognition methods still cannot recognize irregular text well and the recognition accuracy is low, and provide a method for correcting and recognizing irregular text based on a saliency map and system
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0062] In order to make the purpose, features and advantages of the present invention more obvious and understandable, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the following The described embodiments are only some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.
[0063] The technical solutions of the present invention will be further described below in conjunction with the accompanying drawings and through specific implementation methods.
[0064] A saliency graph-based correction and recognition method for irregular text, which is mainly divided into a saliency graph-based irregular text correcti...
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
Login to View More Abstract
The invention discloses an irregular text correction and recognition method and system based on a saliency map, and the method comprises the steps of inputting an irregular text picture into a pre-trained irregular text correction network model, and outputting a character-level classification saliency map; correcting the irregular text picture and the classification saliency map by utilizing a strip-shaped region transformation algorithm, and outputting a corrected picture; and inputting the corrected picture into a pre-trained irregular text recognition network model, and outputting text picture recognition information. The method has the advantages that the irregular text pictures are processed through the trained irregular text correction network model, the strip-shaped region transformation algorithm and the trained irregular text recognition network model, and the accuracy and robustness of irregular text recognition are improved.
Description
technical field [0001] The invention relates to a method and system for correcting and recognizing irregular text based on a saliency map, and belongs to the technical field of natural scene text recognition. Background technique [0002] In recent years, natural scene text recognition has become a research hotspot in computer vision. Its input is a text image, and its output is the text information in the image. Natural scene text recognition has important applications in areas such as automatic driving, image retrieval, document recognition, and container number recognition. Natural scene text recognition can be mainly divided into two categories: regular text recognition and irregular text recognition. Regular text is horizontal text, and irregular text is oblique text, curved text, perspective-distorted text, etc. Irregular text appears in many scenarios, such as bottle caps, trademarks, etc. At present, most text recognition methods can only recognize regular text. F...
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
Login to View More Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/32G06N3/04G06N3/08
CPCG06N3/08G06V10/243G06N3/044G06N3/045Y02D10/00
Inventor 尤晶晶李武军
Owner NANJING UNIV