Efficient power signal description model training method
What is Al technical title?
Al technical title is built by PatSnap Al team. It summarizes the technical point description of the patent document.
A power signal and description model technology, applied in character and pattern recognition, instrument, character recognition, etc., to achieve the effect of improving training efficiency
Pending Publication Date: 2021-07-23
南京太司德智能电气有限公司
View PDF0 Cites 0 Cited by
Summary
Abstract
Description
Claims
Application Information
AI Technical Summary
This helps you quickly interpret patents by identifying the three key elements:
Problems solved by technology
Method used
Benefits of technology
Problems solved by technology
Because the training needs to adjust the position of the text in the picture and the size of the text box a lot, it brings a huge workload to the text training
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
Click on the blue label to locate the original text in one second.
Reading with bidirectional positioning of images and text.
Smart Image
Examples
Experimental program
Comparison scheme
Effect test
Embodiment 1
[0076] Embodiment 1: as figure 1 As shown, an efficient power signal description model training method includes the following steps:
[0077] 1) Form the names of stations and status names in the power dispatching master station system into files in txt format or execl format, and read the txt or excel files; due to the large number of Chinese characters, this step will carry out the text that needs to be recognized in practical applications Sorting, on the one hand, reduces the number of characters to be recognized, and on the other hand, reduces the size of the generated model, which can improve the speed of character recognition;
[0078] 2) Set the parameters of the font, size, and model name of the training text to improve the accuracy of text recognition;
[0079] 3) Read the selected txt or excel file line by line to obtain the total number of lines of text (denoted as num_lines) and the number of words in the line with the largest number of words in these lines (deno...
Embodiment 2
[0131] Embodiment 2: as figure 2 Shown, a method for merging training files, the method includes the following steps:
[0132] 1) selection has been manually marked and has generated a .box text file and a picture file with a suffix of .tif, which can be a .box text file and a suffix of the picture file of .tif generated by steps 1)-10) in embodiment 1, It can also be adjusted by other methods or tools; image files in .tif format under different paths can be selected, which is more convenient than the traditional method of manually copying the merged files to the same path;
[0133] 2) Set the training parameters of each file, including training language and page mode parameters, the default is Chinese training language; different parameters can be set for each selected file, compared with the traditional method, it is necessary to enter parameters and execute commands separately for each file The method is more intuitive and convenient;
[0134] 3) According to the selecte...
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 efficient power signal description model training method. The method comprises the following steps: reading a txt or excel file; setting training character parameters; generating the width and the height of the character picture; judging the width of the picture and recalculating the height of the picture; calling a QImage class of the Qt to generate a full-white picture; drawing a single character on the full-white picture; recording the position of each character and the length and width data of the rectangle; training a character conversion coordinate system; storing the data of each converted character into a text file of which the suffix is. Box; generating a text file of which the suffix is. Tr; reading all files with the suffixes of. Tif,. Box and. Tr in the manually marked folder; executing a training command of tesseract, and generating a file of which the suffix is. Traineddata; and automatically calling a tesseract command to recognize the generated picture, comparing the recognition result with the input characters, and prompting the recognition error characters, the character recall rate and the accuracy rate. According to the invention, the steps of character training are simplified, character training can be quickly realized, the required training model is quickly generated, and the training efficiency is improved.
Description
technical field [0001] The invention belongs to the technical field of Chinese training model training methods, and in particular relates to a highly efficient power signal description model training method. Background technique [0002] The recognition rate of the Chinese training model that comes with the Tesseract text recognition engine is low. It is a common practice for users to improve the text recognition rate by retraining users' frequently used characters. Because the training needs to adjust the position of the text in the picture and the size of the text box a lot, it brings a huge workload to the text training. Contents of the invention [0003] The technical problem to be solved by the present invention is to provide an efficient power signal description model training method to solve the problems existing in the prior art. [0004] The technical scheme that the present invention takes is: a kind of efficient power signal description model training method, t...
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
Application Date:The date an application was filed.
Publication Date:The date a patent or application was officially published.
First Publication Date:The earliest publication date of a patent with the same application number.
Issue Date:Publication date of the patent grant document.
PCT Entry Date:The Entry date of PCT National Phase.
Estimated Expiry Date:The statutory expiry date of a patent right according to the Patent Law, and it is the longest term of protection that the patent right can achieve without the termination of the patent right due to other reasons(Term extension factor has been taken into account ).
Invalid Date:Actual expiry date is based on effective date or publication date of legal transaction data of invalid patent.
Login to view more 
Login to view more