Self-adaptive deviation rectifying method and device, readable medium and electronic device
An adaptive and bias correction technology, applied in the computer field, which can solve problems such as overfitting, inability to obtain bias correction phrases, and inability to accurately correct bias words.
Active Publication Date: 2019-06-07
天津幸福生命科技有限公司 +1
View PDF10 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0004] When the neural network model is used for fitting and correcting the corrected words, it is very easy to overfit, which leads to over-emphasis on the reference meaning in the corrected words during the fitting and corrected process Smaller characters (for example, wrong characters or numbers) cannot get accurate deviation correction phrases, that is, it is impossible to accurately realize deviation correction for deviation correction words
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
[0064] In order to make the purpose, technical solution and advantages of the present invention clearer, the technical solution of the present invention will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Apparently, the described embodiments are only some of the embodiments of the present invention, but not all of them. 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.
[0065] Such as figure 1 As shown, the embodiment of the present invention provides an adaptive deviation correction method, including the following steps:
[0066] S1. Determine the candidate correcting phrase with the greatest similarity to the phrase to be corrected from each preset sample phrase;
[0067] S2. Determine whether the Jaccard coefficient of the phrase to be corrected and...
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 a self-adaptive deviation rectifying method and device, a readable medium and an electronic device. The method comprises: determining a candidate deviation rectification word group with the maximum similarity with a to-be-rectified word group from all preset sample word groups; S2, determining whether the Jacod coefficient of the to-be-rectified phrases and the candidate rectified phrases is greater than a first preset value, if yes, executing S3, and if not, executing S4; S3, determining the candidate correction phrases as correction phrases corresponding to the to-be-corrected phrases; and S4, deleting at least one character of the to-be-rectified phrases, taking the to-be-rectified phrases with the at least one character deleted as the to-be-rectified phrases, and executing the step S1. According to the technical scheme, the to-be-rectified words can be rectified more accurately.
Description
technical field [0001] The invention relates to the technical field of computers, in particular to an adaptive deviation correction method, device, readable medium and electronic equipment. Background technique [0002] Optical Character Recognition (OCR, Optical Character Recognition) is the process of identifying and extracting characters and layout information from image files. The quality of the image file directly affects the accuracy of the recognition result. Therefore, it is usually necessary to rectify the phrase consisting of several characters in the recognition result to obtain a more accurate phrase. [0003] At present, the neural network model is mainly obtained by training a large number of sample phrases, and the trained neural network model is used to fit and rectify the deviation-correcting phrases to obtain corresponding correcting phrases. [0004] When the neural network model is used to fit and correct the corrected words, it is very easy to overfit, ...
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): G06K9/20G06K9/62G06F17/27
Inventor 凤博郭潇宇
Owner 天津幸福生命科技有限公司