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

Industrial radiographic film image lead word number identification method

A recognition method and type technology, which are applied in the field of film character recognition and artificial intelligence, can solve the problems of difficulty in collecting industrial negative type images, and the inability to collect a large number of complete data sets.

Pending Publication Date: 2021-09-14
沈阳派得林科技有限责任公司
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Affected by actual conditions, there are difficulties in the acquisition of industrial film lead image, which makes it impossible to collect a large number of complete data sets

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
  • Industrial radiographic film image lead word number identification method
  • Industrial radiographic film image lead word number identification method
  • Industrial radiographic film image lead word number identification method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0091] Step 1: Extract the beginning part of the image lead number.

[0092] Step 1.1: Dataset labeling. All samples come from real industrial negatives, which are scanned into computer images. Professionals use labeling software to label the data set and generate label files, and each image corresponds to a label file.

[0093] Step 1.2: Image random slice and annotation format conversion.

[0094] Step 1.2.1: The image is randomly sliced.

[0095] The random number is set to 5, that is, 5 small images are generated for each labeled image, and the random step is set to 100. When the random number is less than the random number, set the random number to the random number.

[0096] Step 1.2.2: Annotation format conversion.

[0097] Yolov5 specifies the input annotation format. Each annotation box includes: class, x_center, y_center, width, height, which respectively refer to: sample label, x coordinate of the center point of the annotation box, y coordinate of the center poi...

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 relates to the technical field of negative film character recognition and artificial intelligence, in particular to an industrial ray negative film image lead word number recognition method which is based on deep learning and edge features and comprises a data collection module, an image detection module and a character processing module. According to the method, a fusion algorithm including a traditional edge detection method and a deep learning method is adopted, so that the recognition accuracy can be effectively improved. In order to ensure the validity of the detected lead word number, the algorithm needs to know the beginning part of the lead word number to be detected before identification, so the detection starting point is determined. At the end, a large amount of logic judgment is added in the algorithm to determine the legal end point of the lead word number. Considering that the lead word form change of the bottom film is large and the lead word position has certain offset, corresponding processing is performed in the algorithm, and the tolerance of the algorithm to the original image is ensured. In consideration of image features of an actual negative film, multiple means are adopted for image preprocessing, and the influence of image quality on an algorithm is avoided.

Description

technical field [0001] The invention relates to the technical fields of film character recognition and artificial intelligence, in particular to a method for identifying type numbers of industrial radiographic film images. Background technique [0002] Optical character recognition is an important recognition technology, which can extract the text in the image into computer text, which is convenient for subsequent storage, processing and utilization. Industrial X-ray films are widely used for non-destructive X-ray inspection of device profile parts or welds made of ferrous metals, non-ferrous metals and their alloys or other materials with small decay coefficients. The digitized industrial film image contains a lot of information, such as image number, image shooting date, pipe size, etc. It is necessary to adopt a detection technology to extract this information to assist subsequent film digitization, intelligence, image processing and other tasks. [0003] Existing detect...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/32G06K9/34G06K9/46G06K9/62G06N3/04G06N3/08G06N5/04
CPCG06N5/04G06N3/04G06N3/08G06F18/241G06F18/214Y02P90/30
Inventor 朱宇恒张春娥赵巍王兰
Owner 沈阳派得林科技有限责任公司
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