Unlock instant, AI-driven research and patent intelligence for your innovation.

Seal repair network model training method and device

A technology of network model and training method, which is applied in the field of stamp repair network model training, can solve problems such as poor seal effect, and achieve the effect of improving repair effect, accurate repair, and clear boundaries

Active Publication Date: 2022-08-09
DATACANVAS LTD
View PDF8 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the embodiment of the present invention is to provide a training method and device for a seal repair network model to solve the technical problem that the effect of repairing the seal is not good

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
  • Seal repair network model training method and device
  • Seal repair network model training method and device
  • Seal repair network model training method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0072] The technical solutions in the embodiments of the present invention will be described below with reference to the accompanying drawings in the embodiments of the present invention.

[0073] For ease of understanding, before describing the solution of the present invention in detail, some contents involved in the embodiments of the present application are described:

[0074] The seal repair network model in the embodiment of the present application may be a generative adversarial model (a generative adversarial network model is also referred to as a GAN model). Among them, the generative adversarial network model includes two sets of independent network models, and the two are used as mutual confrontation goals. The first set of network model is the generator, the input of the generator is the noise seal, the generator uses the original seal as the label, repairs the noise seal, and outputs the repaired seal. The second set of network model is the discriminator. The dis...

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 provides a training method and device for a seal repair network model. The training method of the seal repair network model comprises the steps of determining a first target loss value corresponding to a generator according to a repair seal, an original seal, a first edge region seal and a second edge region seal output by the generator; using the first target loss value to update a first network parameter corresponding to the generator; determining a second target loss value corresponding to the discriminator according to the classification result corresponding to the discriminator; and using the second target loss value to update a second network parameter corresponding to the discriminator. In the embodiment of the invention, the first target loss value is determined according to the first marginal area seal and the second marginal area seal, and the first network parameter corresponding to the generator is updated according to the first target loss value, so that the boundary of each character of the repaired seal generated by the generator is clear, and the characters in the repaired seal can be clearly displayed; therefore, the repairing effect is improved.

Description

technical field [0001] The invention relates to the technical field of seal processing, in particular to a training method and device for a seal repairing network model. Background technique [0002] In the field of seal processing technology, the usual way of repairing a seal is to first perform a binarization process on the seal to remove the background noise in the seal; then remove the blurred part of the seal to obtain a repaired seal. [0003] However, when the seal is repaired by the above method, the boundaries between the characters in the repaired seal are not clear enough, and the characters are not displayed clearly. In this way, the repair effect is not good. SUMMARY OF THE INVENTION [0004] The purpose of the embodiments of the present invention is to provide a method and apparatus for training a seal repair network model, so as to solve the technical problem that the effect of repairing the seal is not good. [0005] In a first aspect, an embodiment of th...

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): G06T5/00G06T7/13G06V10/774G06V10/764G06V10/74G06V10/82G06N3/08
CPCG06T7/13G06N3/084G06T2207/20081G06T2207/20084G06T2207/20192G06F18/22G06F18/214G06F18/24G06T5/77
Inventor 方磊石畏徐敏严京旗周审章
Owner DATACANVAS LTD