Supercharge Your Innovation With Domain-Expert AI Agents!

Method and system for generating wine bottle defect samples based on deep neural network

A deep neural network and wine bottle technology, which is applied in the field of defect sample generation, can solve problems such as inability to generate high-efficiency and large-scale wine bottle defect samples, inapplicability to large-scale production of wine bottle surface defect detection, and inability to collect a large number of problems. Generate problems on a large scale, save manpower, and improve generation efficiency

Active Publication Date: 2022-05-13
广州易道智慧信息科技有限公司
View PDF9 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] Taking defect photos through the production line, and then manually attaching them to the 3D model can achieve the purpose of making the defect samples look realistic, but because there are few defect samples in the real wine bottle production environment, it is impossible to collect a large number of them, so it is also necessary Unable to generate high-efficiency and a large number of wine bottle defect samples, it is not suitable for large-scale production of wine bottle surface defect detection

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
  • Method and system for generating wine bottle defect samples based on deep neural network
  • Method and system for generating wine bottle defect samples based on deep neural network
  • Method and system for generating wine bottle defect samples based on deep neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment approach 1

[0077] Before describing the wine bottle defect sample generation method based on deep neural network, the following needs to be explained:

[0078] The wine bottle defects referred to in this application usually include but are not limited to bottle cap defects, label defects, coding defects, bottle body defects, etc.

[0079] In addition, it is also necessary to establish a 3D model of the wine bottle. The establishment of the 3D model of the wine bottle may include the following steps:

[0080] 1) Use Solidworks and other software to create a 1:1 wine bottle 3D model according to the actual size of the wine bottle, and set the model surface to be blank, so as to prepare for the later pasting of surface defect pictures;

[0081] 2) According to the surface size of the bottle cap and body of the wine bottle, design the unfolded dimension drawing of the wine bottle sample; this dimension drawing can be used as the dimension drawing required in step S12 to facilitate the prepro...

Embodiment approach 2

[0152] The applicant found that when the similarity between the second sample representing the 2D wine bottle defect image and the real sample is determined, the second sample can be better evaluated to determine whether it is necessary to optimize the adversarial neural network.

[0153] For this reason, the embodiment of the present application is improved on the basis of the first embodiment, see Figure 4 as shown, Figure 4 The structural diagram of the anti-neural generation network in a deep neural network-based wine bottle defect sample generation method provided for the implementation of the present application, the improvements are as follows:

[0154] The anti-neural generation network also includes a discriminant network structure, which includes a fifth deconvolution kernel, a sixth deconvolution kernel, a seventh deconvolution kernel, and an eighth deconvolution kernel;

[0155] The second feature map and the pre-collected real defect samples used as labels are ...

Embodiment approach 3

[0167] The embodiment of the present application provides a wine bottle defect sample generation system based on a deep neural network, including:

[0168] The obtaining module 301 is used to obtain a data set of real wine bottle defect pictures, perform preprocessing and expansion operations to obtain a sample set, and the sample set includes training samples; it is also used for:

[0169] Processing includes:

[0170] S11. Using the camera to collect images of several defective wine bottles from different angles or directions, and obtain ,in, is the k-th image data of the i-th wine bottle, and both i and k are positive integers.

[0171] S12. Will The data in the group are registered and cut according to the sample size map of the wine bottle, and the obtained ,in Cut out the top of the wine bottle cap, Cut out the circle around the waist of the wine bottle cap, It is the cutout picture of the front of the wine bottle, It is the cutout diagram of the back of ...

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 present application relates to the technical field of defect sample generation, in particular to a method and system for generating defect samples of wine bottles based on a deep neural network. The method includes: obtaining a data set of real wine bottle defect pictures, performing preprocessing and expansion operations to obtain a sample set, inputting the expanded data set into a pre-built adversarial neural network, and inputting the expanded data set into a pre-built adversarial neural network , generating a virtual sample; and inputting it into the generating network structure, generating a virtual sample through the generating network structure; and adaptively attaching the virtual sample to the surface of a pre-built 3D wine bottle model. The defect generation method of the present application outputs a large number of virtual samples of wine bottle defects through the anti-neural generation network, and fits them to the 3D wine bottle model, so as to improve the generation efficiency, stability and authenticity of the wine bottle defect samples.

Description

technical field [0001] The present application relates to the technical field of defect sample generation, in particular to a method and system for generating defect samples of wine bottles based on a deep neural network. Background technique [0002] Virtual samples are mainly used in production line layout and production process logistics simulation. With the rapid development of VR virtual reality technology and digital twin technology, the virtual technology of using computer for production line simulation, prediction and optimization of industrial production system is becoming more and more mature. With the rapid development of digital factory technology, virtual production line simulation technology, as a key technology to verify the stability and availability of physical industrial production lines, has also become increasingly mature. [0003] The wine bottle production line is the key industry of digital twin virtual simulation. There are many problems to be solved...

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 Patents(China)
IPC IPC(8): G06T7/00G06T7/11G06T7/73G06V10/26G06V10/44G06V10/764G06V10/82G06K9/62G06N3/04G06N3/08G06T3/00
CPCG06T7/0004G06T7/11G06T7/73G06T3/0043G06N3/08G06T2207/10004G06T2207/20081G06T2207/20084G06T2207/30108G06N3/045G06F18/241
Inventor 李博郑泽胜
Owner 广州易道智慧信息科技有限公司
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com