Data Intelligent Acquisition Method Based on Computer Vision Image Recognition

Abstract

The invention discloses a data intelligent collection method based on computer vision image recognition, which comprises the following steps of S1 classifying the service recording content, selectingthe numerical value class and the judging class; 2 setting the definition rules of the judgment class and storing the definition rules and the position information in a database; setting the definition rules of the numerical value class and storing definition rules with the location information in database; storing the unique identifier of each product and the location information of the identifier area in the data; S3 marking the results in a marking area of the judgment class and a marking area of the numerical value class, and marking an identifier in the marking area of the identifier; S4carrying out the real-time image acquisition, and recording the time point of image acquisition; S5 carrying out the AI identification on the service record content and the result in the image; S6 converting the tag results into the business data. The method of the invention subverts the thinking mode of data collection of the traditional industry, enhances the digitalized management ability of the enterprise, empowers the traditional industry, accelerates the transformation and upgrading of the traditional industry, and accelerates the practical process of the artificial intelligence.

Description

technical field [0001] The invention belongs to the field of computer software, in particular to an intelligent data collection method based on computer vision image recognition. Background technique [0002] Text recognition technology: Optical character recognition (OCR) is aimed at printed characters, using optical methods to convert the text in paper documents into black and white dot matrix image files, and converting the text in the image into text format through recognition software. Technology for further editing and processing by word processing software. This technology is mainly used to convert printed paper document scanning (such as scanner or digital camera) recognition into computer text and handwritten text recognition, where handwritten text recognition is applied to real-time recognition on mobile phones, tablets and handwriting boards. The technical flaws are as follows: [0003] Industrial and commercial and manufacturing enterprises have a large amount...

AI Technical Summary

Problems solved by technology

[0004] First, paper documents with the same format have different recording tools, resulting in differences in the thickness of text lines, different colors, and depths of handwriting, resulting in low recognition rates;

[0005] Second, the writing habits of different staff members are inconsistent, which leads to the inconsistency in the regularity and scribbling of the recorded text, which makes the text unrecognizable;

[0006] Third, for the paper documents of the same format, different people have different time points, and the starting positions of the records are different. The traditional character recognition technology cannot distinguish the time point and content of the record change, resulting in the inability to accurately record the time when the business occurred and the time of the transaction. Business data recorded in time period

Images